Spaces:
Running
Running
update CRM Bias
Browse files- crm-results/hf_leaderboard_accuracy.csv +45 -45
- crm-results/hf_leaderboard_crm_bias.csv +19 -0
- crm-results/hf_leaderboard_latency_cost.csv +2 -2
- crm-results/hf_leaderboard_ts.csv +3 -3
- src/about.py +2 -2
- src/display/utils.py +2 -1
- src/populate.py +6 -1
crm-results/hf_leaderboard_accuracy.csv
CHANGED
@@ -1,42 +1,42 @@
|
|
1 |
Use Case Name,Use Case Type,Accuracy Method,Model Name,Model Version,LLM Provider,Factuality,Instruction Following,Conciseness,Completeness,Accuracy
|
2 |
-
Service: Conversation summary,Summary,Auto,Gemini
|
3 |
Service: Conversation summary,Summary,Auto,GPT4-o,GPT4-o,OpenAI,3.9669582118561713,3.9961127308066082,3.9300291545189503,3.9844509232264333,3.9693877551020407
|
4 |
Service: Conversation summary,Summary,Auto,AI21 Jamba-Instruct,AI21 (jamba-instruct-preview),AI21,3.9310009718172982,3.998056365403304,3.8104956268221573,3.9737609329446064,3.9283284742468414
|
5 |
Service: Conversation summary,Summary,Auto,GPT 4 Turbo,gpt-4-0613,OpenAI,3.9504373177842567,4.0,3.9067055393586005,3.9805636540330416,3.9594266277939747
|
6 |
Service: Conversation summary,Summary,Auto,SF-TextBase 70B,TextBase-70B (Llama FT),Salesforce,3.9591836734693877,3.998056365403304,3.881438289601555,3.992225461613217,3.957725947521866
|
7 |
Service: Conversation summary,Summary,Auto,Claude 3 Haiku,Claude 3 Haiku,Anthropic,3.9591836734693877,3.999028182701652,3.6997084548104957,3.993197278911565,3.912779397473275
|
8 |
Service: Conversation summary,Summary,Auto,Cohere Command R+,cohere.cmd-R+,Cohere AI,3.954324586977648,4.0,3.8328474246841595,3.9951409135082603,3.945578231292517
|
9 |
-
Service: Conversation summary,Summary,Auto,XGen
|
10 |
Service: Conversation summary,Summary,Auto,SF-TextBase 7B,TextBase-7B (Mistral FT),Salesforce,3.8746355685131197,3.991253644314869,3.7862001943634596,3.9407191448007777,3.898202137998057
|
11 |
Service: Conversation summary,Summary,Auto,LLaMA 3 70B,Meta-Llama-3-70B-Instruct,Meta,3.9591836734693877,3.999028182701652,3.9280855199222544,3.990281827016521,3.9691448007774537
|
12 |
Service: Conversation summary,Summary,Auto,Cohere Command Text,cohere.command-text-v14,Cohere AI,3.7560738581146746,3.938775510204082,3.6530612244897958,3.938775510204082,3.8216715257531586
|
13 |
Service: Conversation summary,Summary,Auto,GPT 3.5 Turbo,gpt-3.5-turbo,OpenAI,3.9310009718172982,3.9961127308066082,3.9834791059280854,3.935860058309038,3.961613216715257
|
14 |
Service: Conversation summary,Summary,Auto,Claude 3 Opus,Claude 3 (Opus),Anthropic,3.9640427599611274,3.998056365403304,3.7657920310981536,3.989310009718173,3.9293002915451893
|
15 |
-
Service: Conversation summary,Summary,Auto,Gemini
|
16 |
Service: Conversation summary,Summary,Auto,LLaMA 3 8B,Meta-Llama-3-8B-Instruct,Meta,3.922254616132167,4.0,3.565597667638484,3.9961127308066082,3.8709912536443145
|
17 |
Service: Conversation summary,Summary,Auto,Mistral 7B,Mistral-7B-Instruct-v0.1,Mistral,3.7764820213799806,3.927113702623907,3.327502429543246,3.924198250728863,3.738824101068999
|
18 |
Service: Conversation summary,Summary,Auto,Mixtral 8x7B,Mixtral-8x7B-v0.1,Mistral,3.9115646258503403,3.987366375121477,3.3751214771622933,3.9825072886297375,3.814139941690962
|
19 |
Service: Conversation summary,Summary,Auto,SF-TextSum,Summarization model 7B for Service (Mistral FT),Salesforce,3.9037900874635567,3.997084548104956,3.7755102040816326,3.9280855199222544,3.9011175898931
|
20 |
Sales: Email Generation,Generation,Auto,Mistral 7B,Mistral-7B-Instruct-v0.1,Mistral,2.2452830188679247,1.9811320754716981,1.8867924528301887,2.018867924528302,2.0330188679245285
|
21 |
-
Sales: Email Generation,Generation,Auto,Gemini
|
22 |
Sales: Email Generation,Generation,Auto,LLaMA 3 8B,Meta-Llama-3-8B-Instruct,Meta,3.660377358490566,3.7358490566037736,3.5849056603773586,3.0754716981132075,3.5141509433962264
|
23 |
Sales: Email Generation,Generation,Auto,SF-TextSum,Summarization model 7B for Service (Mistral FT),Salesforce,3.4716981132075473,3.4150943396226414,3.5283018867924527,2.9245283018867925,3.334905660377359
|
24 |
Sales: Email Generation,Generation,Auto,Claude 3 Haiku,Claude 3 Haiku,Anthropic,3.0377358490566038,2.5283018867924527,2.5283018867924527,2.811320754716981,2.7264150943396226
|
25 |
Sales: Email Generation,Generation,Auto,Cohere Command R+,cohere.cmd-R+,Cohere AI,3.7358490566037736,3.6792452830188678,3.547169811320755,3.018867924528302,3.4952830188679247
|
26 |
Sales: Email Generation,Generation,Auto,Cohere Command Text,cohere.command-text-v14,Cohere AI,2.7547169811320753,2.358490566037736,2.2641509433962264,2.452830188679245,2.4575471698113205
|
27 |
-
Sales: Email Generation,Generation,Auto,XGen
|
28 |
Sales: Email Generation,Generation,Auto,SF-TextBase 70B,TextBase-70B (Llama FT),Salesforce,3.811320754716981,3.849056603773585,3.69811320754717,3.018867924528302,3.5943396226415096
|
29 |
Sales: Email Generation,Generation,Auto,LLaMA 3 70B,Meta-Llama-3-70B-Instruct,Meta,3.811320754716981,3.849056603773585,3.7358490566037736,3.0,3.599056603773585
|
30 |
Sales: Email Generation,Generation,Auto,Claude 3 Opus,Claude 3 (Opus),Anthropic,3.452830188679245,3.056603773584906,3.509433962264151,3.0,3.2547169811320753
|
31 |
Sales: Email Generation,Generation,Auto,GPT4-o,GPT4-o,OpenAI,3.792452830188679,3.830188679245283,3.4716981132075473,3.0377358490566038,3.533018867924528
|
32 |
Sales: Email Generation,Generation,Auto,Mixtral 8x7B,Mixtral-8x7B-v0.1,Mistral,3.0377358490566038,2.9245283018867925,2.7547169811320753,2.69811320754717,2.8537735849056607
|
33 |
-
Sales: Email Generation,Generation,Auto,Gemini
|
34 |
Sales: Email Generation,Generation,Auto,GPT 3.5 Turbo,gpt-3.5-turbo,OpenAI,3.4339622641509435,3.056603773584906,3.056603773584906,2.69811320754717,3.061320754716981
|
35 |
Sales: Email Generation,Generation,Auto,GPT 4 Turbo,gpt-4-0613,OpenAI,3.7358490566037736,3.547169811320755,3.30188679245283,3.0,3.3962264150943398
|
36 |
Sales: Email Generation,Generation,Auto,AI21 Jamba-Instruct,AI21 (jamba-instruct-preview),AI21,2.6226415094339623,2.169811320754717,2.150943396226415,2.490566037735849,2.3584905660377355
|
37 |
Sales: Email Generation,Generation,Auto,SF-TextBase 7B,TextBase-7B (Mistral FT),Salesforce,3.4716981132075473,3.150943396226415,3.150943396226415,2.8867924528301887,3.165094339622642
|
38 |
Service: Reply Recommendations,Generation,Auto,Mistral 7B,Mistral-7B-Instruct-v0.1,Mistral,2.7777777777777777,3.3333333333333335,2.9444444444444446,2.7222222222222223,2.9444444444444446
|
39 |
-
Service: Reply Recommendations,Generation,Auto,Gemini
|
40 |
Service: Reply Recommendations,Generation,Auto,GPT4-o,GPT4-o,OpenAI,3.388888888888889,3.9444444444444446,3.8333333333333335,3.111111111111111,3.5694444444444446
|
41 |
Service: Reply Recommendations,Generation,Auto,SF-TextBase 70B,TextBase-70B (Llama FT),Salesforce,3.2222222222222223,3.9444444444444446,3.6666666666666665,3.3333333333333335,3.541666666666667
|
42 |
Service: Reply Recommendations,Generation,Auto,LLaMA 3 8B,Meta-Llama-3-8B-Instruct,Meta,3.111111111111111,3.9444444444444446,3.5555555555555554,3.111111111111111,3.4305555555555554
|
@@ -44,107 +44,107 @@ Service: Reply Recommendations,Generation,Auto,Claude 3 Haiku,Claude 3 Haiku,Ant
|
|
44 |
Service: Reply Recommendations,Generation,Auto,Cohere Command R+,cohere.cmd-R+,Cohere AI,3.0,3.611111111111111,2.888888888888889,3.1666666666666665,3.1666666666666665
|
45 |
Service: Reply Recommendations,Generation,Auto,Cohere Command Text,cohere.command-text-v14,Cohere AI,3.0555555555555554,3.2777777777777777,2.5555555555555554,3.1666666666666665,3.013888888888889
|
46 |
Service: Reply Recommendations,Generation,Auto,SF-TextBase 7B,TextBase-7B (Mistral FT),Salesforce,3.2777777777777777,3.5555555555555554,3.111111111111111,2.8333333333333335,3.1944444444444446
|
47 |
-
Service: Reply Recommendations,Generation,Auto,XGen
|
48 |
Service: Reply Recommendations,Generation,Auto,LLaMA 3 70B,Meta-Llama-3-70B-Instruct,Meta,3.1666666666666665,4.0,3.888888888888889,3.2777777777777777,3.583333333333333
|
49 |
Service: Reply Recommendations,Generation,Auto,Claude 3 Opus,Claude 3 (Opus),Anthropic,3.3333333333333335,4.0,3.388888888888889,3.5555555555555554,3.5694444444444446
|
50 |
Service: Reply Recommendations,Generation,Auto,Mixtral 8x7B,Mixtral-8x7B-v0.1,Mistral,3.2222222222222223,3.888888888888889,3.611111111111111,3.1666666666666665,3.4722222222222223
|
51 |
-
Service: Reply Recommendations,Generation,Auto,Gemini
|
52 |
Service: Reply Recommendations,Generation,Auto,GPT 3.5 Turbo,gpt-3.5-turbo,OpenAI,3.1666666666666665,3.888888888888889,3.7777777777777777,3.0555555555555554,3.472222222222222
|
53 |
Service: Reply Recommendations,Generation,Auto,GPT 4 Turbo,gpt-4-0613,OpenAI,3.2222222222222223,3.9444444444444446,3.5,3.1666666666666665,3.458333333333333
|
54 |
Service: Reply Recommendations,Generation,Auto,AI21 Jamba-Instruct,AI21 (jamba-instruct-preview),AI21,3.0,3.8333333333333335,3.2777777777777777,3.2222222222222223,3.333333333333333
|
55 |
Service: Reply Recommendations,Generation,Auto,SF-TextSum,Summarization model 7B for Service (Mistral FT),Salesforce,3.2777777777777777,3.888888888888889,3.2777777777777777,3.1666666666666665,3.4027777777777777
|
56 |
-
Sales & Service: Update CRM Info,Generation,Auto,Gemini
|
57 |
Sales & Service: Update CRM Info,Generation,Auto,GPT4-o,GPT4-o,OpenAI,3.909090909090909,3.8848484848484848,3.915151515151515,3.6484848484848484,3.83939393939394
|
58 |
Sales & Service: Update CRM Info,Generation,Auto,AI21 Jamba-Instruct,AI21 (jamba-instruct-preview),AI21,3.6787878787878787,3.581818181818182,3.709090909090909,3.533333333333333,3.625757575757576
|
59 |
Sales & Service: Update CRM Info,Generation,Auto,GPT 4 Turbo,gpt-4-0613,OpenAI,3.7333333333333334,3.709090909090909,3.806060606060606,3.6363636363636362,3.7212121212121216
|
60 |
Sales & Service: Update CRM Info,Generation,Auto,SF-TextBase 70B,TextBase-70B (Llama FT),Salesforce,3.806060606060606,3.824242424242424,3.83030303030303,3.775757575757576,3.809090909090909
|
61 |
Sales & Service: Update CRM Info,Generation,Auto,Claude 3 Haiku,Claude 3 Haiku,Anthropic,3.6666666666666665,3.618181818181818,3.7151515151515153,3.624242424242424,3.656060606060606
|
62 |
Sales & Service: Update CRM Info,Generation,Auto,Cohere Command R+,cohere.cmd-R+,Cohere AI,3.812121212121212,3.7636363636363637,3.757575757575758,3.624242424242424,3.7393939393939393
|
63 |
-
Sales & Service: Update CRM Info,Generation,Auto,XGen
|
64 |
Sales & Service: Update CRM Info,Generation,Auto,SF-TextBase 7B,TextBase-7B (Mistral FT),Salesforce,3.096969696969697,2.9696969696969697,2.8,3.0545454545454547,2.9803030303030305
|
65 |
Sales & Service: Update CRM Info,Generation,Auto,LLaMA 3 70B,Meta-Llama-3-70B-Instruct,Meta,3.812121212121212,3.806060606060606,3.8181818181818183,3.7515151515151515,3.796969696969697
|
66 |
Sales & Service: Update CRM Info,Generation,Auto,Cohere Command Text,cohere.command-text-v14,Cohere AI,2.903030303030303,2.909090909090909,2.8666666666666667,2.896969696969697,2.893939393939394
|
67 |
Sales & Service: Update CRM Info,Generation,Auto,GPT 3.5 Turbo,gpt-3.5-turbo,OpenAI,3.7515151515151515,3.703030303030303,3.7515151515151515,3.6606060606060606,3.716666666666667
|
68 |
Sales & Service: Update CRM Info,Generation,Auto,Claude 3 Opus,Claude 3 (Opus),Anthropic,3.7636363636363637,3.7454545454545456,3.812121212121212,3.606060606060606,3.731818181818182
|
69 |
-
Sales & Service: Update CRM Info,Generation,Auto,Gemini
|
70 |
Sales & Service: Update CRM Info,Generation,Auto,LLaMA 3 8B,Meta-Llama-3-8B-Instruct,Meta,3.3575757575757574,2.787878787878788,2.3333333333333335,3.242424242424242,2.93030303030303
|
71 |
Sales & Service: Update CRM Info,Generation,Auto,Mistral 7B,Mistral-7B-Instruct-v0.1,Mistral,1.6787878787878787,1.6606060606060606,1.6424242424242423,1.6848484848484848,1.6666666666666667
|
72 |
Sales & Service: Update CRM Info,Generation,Auto,Mixtral 8x7B,Mixtral-8x7B-v0.1,Mistral,3.690909090909091,3.6666666666666665,3.690909090909091,3.6363636363636362,3.6712121212121214
|
73 |
Sales & Service: Update CRM Info,Generation,Auto,SF-TextSum,Summarization model 7B for Service (Mistral FT),Salesforce,3.1575757575757577,2.981818181818182,3.012121212121212,3.139393939393939,3.0727272727272728
|
74 |
-
Service: Live Chat Summary,Summary,Auto,Gemini
|
75 |
Service: Live Chat Summary,Summary,Auto,GPT4-o,GPT4-o,OpenAI,3.641,3.965,3.946,3.436,3.747
|
76 |
Service: Live Chat Summary,Summary,Auto,AI21 Jamba-Instruct,AI21 (jamba-instruct-preview),AI21,3.6236236236236237,3.715,3.714,3.331,3.595905905905906
|
77 |
Service: Live Chat Summary,Summary,Auto,SF-TextBase 70B,TextBase-70B (Llama FT),Salesforce,3.718,3.882,3.904,3.4004004004004003,3.7261001001001
|
78 |
Service: Live Chat Summary,Summary,Auto,Claude 3 Haiku,Claude 3 Haiku,Anthropic,3.588,3.927,3.607,3.566,3.6720000000000006
|
79 |
Service: Live Chat Summary,Summary,Auto,Cohere Command R+,cohere.cmd-R+,Cohere AI,3.453,3.945,3.666,3.725,3.69725
|
80 |
-
Service: Live Chat Summary,Summary,Auto,XGen
|
81 |
Service: Live Chat Summary,Summary,Auto,SF-TextBase 7B,TextBase-7B (Mistral FT),Salesforce,3.593,3.694,3.75,3.158,3.54875
|
82 |
Service: Live Chat Summary,Summary,Auto,LLaMA 3 70B,Meta-Llama-3-70B-Instruct,Meta,3.765,3.903,3.947,3.309,3.731
|
83 |
Service: Live Chat Summary,Summary,Auto,Cohere Command Text,cohere.command-text-v14,Cohere AI,3.502,3.744,3.533,3.497,3.569
|
84 |
Service: Live Chat Summary,Summary,Auto,GPT 4 Turbo,gpt-4-0613,OpenAI,3.587,3.992,3.914,3.822,3.82875
|
85 |
Service: Live Chat Summary,Summary,Auto,GPT 3.5 Turbo,gpt-3.5-turbo,OpenAI,3.717,3.835,3.891,3.164164164164164,3.651791041041041
|
86 |
Service: Live Chat Summary,Summary,Auto,Claude 3 Opus,Claude 3 (Opus),Anthropic,3.574,3.862,3.864,3.492,3.698
|
87 |
-
Service: Live Chat Summary,Summary,Auto,Gemini
|
88 |
Service: Live Chat Summary,Summary,Auto,LLaMA 3 8B,Meta-Llama-3-8B-Instruct,Meta,3.681,3.93,3.772,3.7,3.7707500000000005
|
89 |
Service: Live Chat Summary,Summary,Auto,Mistral 7B,Mistral-7B-Instruct-v0.1,Mistral,3.492,3.942,3.831,3.353,3.6544999999999996
|
90 |
Service: Live Chat Summary,Summary,Auto,Mixtral 8x7B,Mixtral-8x7B-v0.1,Mistral,3.609,3.866,3.846,3.379,3.6750000000000003
|
91 |
Service: Live Chat Summary,Summary,Auto,SF-TextSum,Summarization model 7B for Service (Mistral FT),Salesforce,3.504,3.595,3.709,3.129129129129129,3.4842822822822823
|
92 |
-
Service: Email Summary,Summary,Auto,Gemini
|
93 |
Service: Email Summary,Summary,Auto,GPT4-o,GPT4-o,OpenAI,3.752577319587629,3.9381443298969074,3.8762886597938144,3.9587628865979383,3.8814432989690726
|
94 |
Service: Email Summary,Summary,Auto,AI21 Jamba-Instruct,AI21 (jamba-instruct-preview),AI21,3.377551020408163,3.6122448979591835,3.4285714285714284,3.5306122448979593,3.4872448979591835
|
95 |
Service: Email Summary,Summary,Auto,SF-TextBase 70B,TextBase-70B (Llama FT),Salesforce,3.673469387755102,3.989795918367347,3.9693877551020407,3.979591836734694,3.9030612244897958
|
96 |
Service: Email Summary,Summary,Auto,Claude 3 Haiku,Claude 3 Haiku,Anthropic,3.857142857142857,3.979591836734694,3.673469387755102,3.9591836734693877,3.86734693877551
|
97 |
Service: Email Summary,Summary,Auto,Cohere Command R+,cohere.cmd-R+,Cohere AI,3.693877551020408,4.0,3.857142857142857,4.0,3.8877551020408165
|
98 |
-
Service: Email Summary,Summary,Auto,XGen
|
99 |
Service: Email Summary,Summary,Auto,SF-TextBase 7B,TextBase-7B (Mistral FT),Salesforce,3.479591836734694,3.7755102040816326,3.5816326530612246,3.8979591836734695,3.683673469387755
|
100 |
Service: Email Summary,Summary,Auto,LLaMA 3 70B,Meta-Llama-3-70B-Instruct,Meta,3.622448979591837,4.0,3.979591836734694,3.989795918367347,3.8979591836734695
|
101 |
Service: Email Summary,Summary,Auto,Cohere Command Text,cohere.command-text-v14,Cohere AI,3.3877551020408165,3.9591836734693877,3.7448979591836733,3.836734693877551,3.732142857142857
|
102 |
Service: Email Summary,Summary,Auto,GPT 4 Turbo,gpt-4-0613,OpenAI,3.795918367346939,3.989795918367347,3.8469387755102042,4.0,3.908163265306123
|
103 |
Service: Email Summary,Summary,Auto,GPT 3.5 Turbo,gpt-3.5-turbo,OpenAI,3.5918367346938775,3.989795918367347,3.9693877551020407,3.86734693877551,3.854591836734694
|
104 |
Service: Email Summary,Summary,Auto,Claude 3 Opus,Claude 3 (Opus),Anthropic,3.795918367346939,4.0,3.8469387755102042,3.979591836734694,3.9056122448979593
|
105 |
-
Service: Email Summary,Summary,Auto,Gemini
|
106 |
Service: Email Summary,Summary,Auto,LLaMA 3 8B,Meta-Llama-3-8B-Instruct,Meta,3.693877551020408,4.0,3.86734693877551,3.979591836734694,3.885204081632653
|
107 |
Service: Email Summary,Summary,Auto,Mistral 7B,Mistral-7B-Instruct-v0.1,Mistral,3.642857142857143,3.8877551020408165,3.520408163265306,3.8979591836734695,3.737244897959184
|
108 |
Service: Email Summary,Summary,Auto,Mixtral 8x7B,Mixtral-8x7B-v0.1,Mistral,3.673469387755102,3.979591836734694,3.7346938775510203,3.9591836734693877,3.836734693877551
|
109 |
Service: Email Summary,Summary,Auto,SF-TextSum,Summarization model 7B for Service (Mistral FT),Salesforce,3.377551020408163,3.377551020408163,3.36734693877551,3.6530612244897958,3.443877551020408
|
110 |
-
Service: Knowledge creation from Case Info,Generation,Auto,Gemini
|
111 |
Service: Knowledge creation from Case Info,Generation,Auto,GPT4-o,GPT4-o,OpenAI,3.3125,3.625,3.1875,3.6875,3.453125
|
112 |
Service: Knowledge creation from Case Info,Generation,Auto,AI21 Jamba-Instruct,AI21 (jamba-instruct-preview),AI21,3.125,3.125,2.8125,3.125,3.046875
|
113 |
Service: Knowledge creation from Case Info,Generation,Auto,SF-TextBase 70B,TextBase-70B (Llama FT),Salesforce,3.1875,3.5625,3.25,3.4375,3.359375
|
114 |
Service: Knowledge creation from Case Info,Generation,Auto,Claude 3 Haiku,Claude 3 Haiku,Anthropic,3.125,3.5625,3.1875,3.25,3.28125
|
115 |
Service: Knowledge creation from Case Info,Generation,Auto,Cohere Command R+,cohere.cmd-R+,Cohere AI,3.0,3.5625,3.125,3.1875,3.21875
|
116 |
-
Service: Knowledge creation from Case Info,Generation,Auto,XGen
|
117 |
Service: Knowledge creation from Case Info,Generation,Auto,SF-TextBase 7B,TextBase-7B (Mistral FT),Salesforce,3.3125,3.5625,3.0625,3.375,3.328125
|
118 |
Service: Knowledge creation from Case Info,Generation,Auto,LLaMA 3 70B,Meta-Llama-3-70B-Instruct,Meta,3.0625,2.4375,2.0625,2.3125,2.46875
|
119 |
Service: Knowledge creation from Case Info,Generation,Auto,Cohere Command Text,cohere.command-text-v14,Cohere AI,3.125,3.0625,2.8125,3.0625,3.015625
|
120 |
Service: Knowledge creation from Case Info,Generation,Auto,GPT 4 Turbo,gpt-4-0613,OpenAI,3.3125,3.6875,3.1875,3.375,3.390625
|
121 |
Service: Knowledge creation from Case Info,Generation,Auto,GPT 3.5 Turbo,gpt-3.5-turbo,OpenAI,3.25,3.5625,3.3125,3.4375,3.390625
|
122 |
Service: Knowledge creation from Case Info,Generation,Auto,Claude 3 Opus,Claude 3 (Opus),Anthropic,3.125,3.375,2.9375,3.1875,3.15625
|
123 |
-
Service: Knowledge creation from Case Info,Generation,Auto,Gemini
|
124 |
Service: Knowledge creation from Case Info,Generation,Auto,LLaMA 3 8B,Meta-Llama-3-8B-Instruct,Meta,3.1875,3.5625,3.0,3.4375,3.296875
|
125 |
Service: Knowledge creation from Case Info,Generation,Auto,Mistral 7B,Mistral-7B-Instruct-v0.1,Mistral,3.0625,3.5625,2.9375,3.375,3.234375
|
126 |
Service: Knowledge creation from Case Info,Generation,Auto,Mixtral 8x7B,Mixtral-8x7B-v0.1,Mistral,3.25,3.4375,3.125,3.375,3.296875
|
127 |
Service: Knowledge creation from Case Info,Generation,Auto,SF-TextSum,Summarization model 7B for Service (Mistral FT),Salesforce,3.125,3.4375,3.0,3.3125,3.21875
|
128 |
-
Sales: Email Summary,Summary,Auto,Gemini
|
129 |
Sales: Email Summary,Summary,Auto,GPT4-o,GPT4-o,OpenAI,3.8877551020408165,3.9693877551020407,3.86734693877551,3.8979591836734695,3.9056122448979593
|
130 |
Sales: Email Summary,Summary,Auto,AI21 Jamba-Instruct,AI21 (jamba-instruct-preview),AI21,3.5714285714285716,3.8979591836734695,3.9183673469387754,3.663265306122449,3.7627551020408165
|
131 |
Sales: Email Summary,Summary,Auto,SF-TextBase 70B,TextBase-70B (Llama FT),Salesforce,3.663265306122449,3.979591836734694,3.9591836734693877,3.836734693877551,3.8596938775510203
|
132 |
Sales: Email Summary,Summary,Auto,Claude 3 Haiku,Claude 3 Haiku,Anthropic,3.857142857142857,3.989795918367347,3.9183673469387754,3.7653061224489797,3.88265306122449
|
133 |
Sales: Email Summary,Summary,Auto,Cohere Command R+,cohere.cmd-R+,Cohere AI,3.7244897959183674,3.8979591836734695,3.795918367346939,3.7551020408163267,3.793367346938776
|
134 |
-
Sales: Email Summary,Summary,Auto,XGen
|
135 |
Sales: Email Summary,Summary,Auto,SF-TextBase 7B,TextBase-7B (Mistral FT),Salesforce,3.357142857142857,3.6020408163265305,3.663265306122449,3.510204081632653,3.5331632653061225
|
136 |
Sales: Email Summary,Summary,Auto,LLaMA 3 70B,Meta-Llama-3-70B-Instruct,Meta,3.642857142857143,3.9693877551020407,3.9591836734693877,3.8979591836734695,3.86734693877551
|
137 |
Sales: Email Summary,Summary,Auto,Cohere Command Text,cohere.command-text-v14,Cohere AI,3.154639175257732,2.9690721649484537,2.649484536082474,3.2783505154639174,3.0128865979381443
|
138 |
Sales: Email Summary,Summary,Auto,GPT 4 Turbo,gpt-4-0613,OpenAI,3.826530612244898,4.0,3.9285714285714284,3.9591836734693877,3.928571428571429
|
139 |
Sales: Email Summary,Summary,Auto,GPT 3.5 Turbo,gpt-3.5-turbo,OpenAI,3.7857142857142856,3.9591836734693877,3.9693877551020407,3.693877551020408,3.8520408163265305
|
140 |
Sales: Email Summary,Summary,Auto,Claude 3 Opus,Claude 3 (Opus),Anthropic,3.826530612244898,4.0,3.9183673469387754,3.938775510204082,3.920918367346939
|
141 |
-
Sales: Email Summary,Summary,Auto,Gemini
|
142 |
Sales: Email Summary,Summary,Auto,LLaMA 3 8B,Meta-Llama-3-8B-Instruct,Meta,3.7346938775510203,4.0,3.9693877551020407,3.806122448979592,3.877551020408163
|
143 |
Sales: Email Summary,Summary,Auto,Mistral 7B,Mistral-7B-Instruct-v0.1,Mistral,3.6288659793814433,3.948453608247423,3.7653061224489797,3.948453608247423,3.822769829581317
|
144 |
Sales: Email Summary,Summary,Auto,Mixtral 8x7B,Mixtral-8x7B-v0.1,Mistral,3.5306122448979593,3.9591836734693877,3.8469387755102042,3.816326530612245,3.788265306122449
|
145 |
Sales: Email Summary,Summary,Auto,SF-TextSum,Summarization model 7B for Service (Mistral FT),Salesforce,3.5306122448979593,3.836734693877551,3.7346938775510203,3.642857142857143,3.686224489795918
|
146 |
Service: Call Summary,Summary,Auto,Mistral 7B,Mistral-7B-Instruct-v0.1,Mistral,3.75,3.95,3.925,3.925,3.8875
|
147 |
-
Service: Call Summary,Summary,Auto,Gemini
|
148 |
Service: Call Summary,Summary,Auto,GPT4-o,GPT4-o,OpenAI,3.925,3.95,3.95,3.925,3.9375
|
149 |
Service: Call Summary,Summary,Auto,SF-TextBase 70B,TextBase-70B (Llama FT),Salesforce,3.975,3.95,3.95,3.95,3.9562500000000003
|
150 |
Service: Call Summary,Summary,Auto,LLaMA 3 8B,Meta-Llama-3-8B-Instruct,Meta,3.9,3.95,3.95,3.95,3.9375000000000004
|
@@ -152,47 +152,47 @@ Service: Call Summary,Summary,Auto,Claude 3 Haiku,Claude 3 Haiku,Anthropic,3.8,3
|
|
152 |
Service: Call Summary,Summary,Auto,Cohere Command R+,cohere.cmd-R+,Cohere AI,3.925,3.95,3.95,3.925,3.9375
|
153 |
Service: Call Summary,Summary,Auto,Cohere Command Text,cohere.command-text-v14,Cohere AI,3.6,3.75,3.675,3.85,3.7187499999999996
|
154 |
Service: Call Summary,Summary,Auto,SF-TextBase 7B,TextBase-7B (Mistral FT),Salesforce,3.95,4.0,3.975,3.875,3.95
|
155 |
-
Service: Call Summary,Summary,Auto,XGen
|
156 |
Service: Call Summary,Summary,Auto,LLaMA 3 70B,Meta-Llama-3-70B-Instruct,Meta,3.875,3.95,3.95,3.925,3.925
|
157 |
Service: Call Summary,Summary,Auto,Claude 3 Opus,Claude 3 (Opus),Anthropic,3.95,4.0,4.0,4.0,3.9875
|
158 |
Service: Call Summary,Summary,Auto,Mixtral 8x7B,Mixtral-8x7B-v0.1,Mistral,3.825,3.975,3.95,3.925,3.91875
|
159 |
-
Service: Call Summary,Summary,Auto,Gemini
|
160 |
Service: Call Summary,Summary,Auto,GPT 3.5 Turbo,gpt-3.5-turbo,OpenAI,3.85,3.95,3.95,3.875,3.90625
|
161 |
Service: Call Summary,Summary,Auto,GPT 4 Turbo,gpt-4-0613,OpenAI,3.9,3.95,3.95,3.925,3.93125
|
162 |
Service: Call Summary,Summary,Auto,AI21 Jamba-Instruct,AI21 (jamba-instruct-preview),AI21,3.825,3.95,3.925,3.9,3.8999999999999995
|
163 |
Service: Call Summary,Summary,Auto,SF-TextSum,Summarization model 7B for Service (Mistral FT),Salesforce,3.875,3.95,3.95,3.875,3.9125
|
164 |
-
Sales: Call Summary,Summary,Auto,Gemini
|
165 |
Sales: Call Summary,Summary,Auto,GPT4-o,GPT4-o,OpenAI,3.9473684210526314,3.9473684210526314,3.473684210526316,3.210526315789474,3.6447368421052633
|
166 |
Sales: Call Summary,Summary,Auto,AI21 Jamba-Instruct,AI21 (jamba-instruct-preview),AI21,3.3684210526315788,3.3157894736842106,3.1052631578947367,3.0,3.1973684210526314
|
167 |
Sales: Call Summary,Summary,Auto,SF-TextBase 70B,TextBase-70B (Llama FT),Salesforce,3.6842105263157894,3.3684210526315788,3.263157894736842,2.789473684210526,3.276315789473684
|
168 |
Sales: Call Summary,Summary,Auto,Claude 3 Haiku,Claude 3 Haiku,Anthropic,3.6315789473684212,3.473684210526316,3.3157894736842106,3.0526315789473686,3.368421052631579
|
169 |
Sales: Call Summary,Summary,Auto,Cohere Command R+,cohere.cmd-R+,Cohere AI,3.789473684210526,3.6842105263157894,2.8421052631578947,3.0526315789473686,3.3421052631578942
|
170 |
-
Sales: Call Summary,Summary,Auto,XGen
|
171 |
Sales: Call Summary,Summary,Auto,SF-TextBase 7B,TextBase-7B (Mistral FT),Salesforce,3.210526315789474,3.1052631578947367,3.210526315789474,2.6315789473684212,3.0394736842105265
|
172 |
Sales: Call Summary,Summary,Auto,LLaMA 3 70B,Meta-Llama-3-70B-Instruct,Meta,3.4210526315789473,3.263157894736842,3.473684210526316,2.8421052631578947,3.2499999999999996
|
173 |
Sales: Call Summary,Summary,Auto,Cohere Command Text,cohere.command-text-v14,Cohere AI,1.7894736842105263,2.473684210526316,2.1052631578947367,2.3157894736842106,2.1710526315789473
|
174 |
Sales: Call Summary,Summary,Auto,GPT 4 Turbo,gpt-4-0613,OpenAI,4.0,3.736842105263158,3.3157894736842106,3.0,3.513157894736842
|
175 |
Sales: Call Summary,Summary,Auto,GPT 3.5 Turbo,gpt-3.5-turbo,OpenAI,3.8421052631578947,3.3684210526315788,3.8421052631578947,2.888888888888889,3.485380116959064
|
176 |
Sales: Call Summary,Summary,Auto,Claude 3 Opus,Claude 3 (Opus),Anthropic,3.9473684210526314,3.8947368421052633,3.5789473684210527,3.0,3.605263157894737
|
177 |
-
Sales: Call Summary,Summary,Auto,Gemini
|
178 |
Sales: Call Summary,Summary,Auto,LLaMA 3 8B,Meta-Llama-3-8B-Instruct,Meta,2.9473684210526314,3.1578947368421053,2.6315789473684212,2.8947368421052633,2.9078947368421053
|
179 |
Sales: Call Summary,Summary,Auto,Mistral 7B,Mistral-7B-Instruct-v0.1,Mistral,3.5555555555555554,3.4210526315789473,3.210526315789474,3.1052631578947367,3.323099415204678
|
180 |
Sales: Call Summary,Summary,Auto,Mixtral 8x7B,Mixtral-8x7B-v0.1,Mistral,3.263157894736842,3.263157894736842,2.6315789473684212,3.0,3.0394736842105265
|
181 |
Sales: Call Summary,Summary,Auto,SF-TextSum,Summarization model 7B for Service (Mistral FT),Salesforce,3.0526315789473686,2.9473684210526314,3.0,2.736842105263158,2.9342105263157894
|
182 |
-
Service: Live Chat Insights,Summary,Auto,Gemini
|
183 |
Service: Live Chat Insights,Summary,Auto,GPT4-o,GPT4-o,OpenAI,3.91015625,4.0,3.9765625,3.91796875,3.951171875
|
184 |
Service: Live Chat Insights,Summary,Auto,AI21 Jamba-Instruct,AI21 (jamba-instruct-preview),AI21,3.859375,3.97265625,3.8515625,3.8203125,3.8759765625
|
185 |
Service: Live Chat Insights,Summary,Auto,SF-TextBase 70B,TextBase-70B (Llama FT),Salesforce,3.85546875,3.99609375,3.94140625,3.96875,3.9404296875
|
186 |
Service: Live Chat Insights,Summary,Auto,Claude 3 Haiku,Claude 3 Haiku,Anthropic,3.8828125,4.0,3.796875,3.98046875,3.9150390625
|
187 |
Service: Live Chat Insights,Summary,Auto,Cohere Command R+,cohere.cmd-R+,Cohere AI,3.87109375,3.984375,3.8671875,3.9375,3.9150390625
|
188 |
-
Service: Live Chat Insights,Summary,Auto,XGen
|
189 |
Service: Live Chat Insights,Summary,Auto,SF-TextBase 7B,TextBase-7B (Mistral FT),Salesforce,3.71484375,3.97265625,3.90625,3.6875,3.8203125
|
190 |
Service: Live Chat Insights,Summary,Auto,LLaMA 3 70B,Meta-Llama-3-70B-Instruct,Meta,3.85546875,3.9921875,3.97265625,3.93359375,3.9384765625
|
191 |
Service: Live Chat Insights,Summary,Auto,Cohere Command Text,cohere.command-text-v14,Cohere AI,3.51171875,3.9375,3.484375,3.8671875,3.7001953125
|
192 |
Service: Live Chat Insights,Summary,Auto,GPT 4 Turbo,gpt-4-0613,OpenAI,3.91015625,3.9921875,3.93359375,3.9375,3.943359375
|
193 |
Service: Live Chat Insights,Summary,Auto,GPT 3.5 Turbo,gpt-3.5-turbo,OpenAI,3.90234375,3.96875,4.0,3.71875,3.8974609375
|
194 |
Service: Live Chat Insights,Summary,Auto,Claude 3 Opus,Claude 3 (Opus),Anthropic,3.9140625,4.0,3.82421875,3.98046875,3.9296875
|
195 |
-
Service: Live Chat Insights,Summary,Auto,Gemini
|
196 |
Service: Live Chat Insights,Summary,Auto,LLaMA 3 8B,Meta-Llama-3-8B-Instruct,Meta,3.8359375,3.9921875,3.62890625,3.96484375,3.85546875
|
197 |
Service: Live Chat Insights,Summary,Auto,Mistral 7B,Mistral-7B-Instruct-v0.1,Mistral,3.8125,3.9921875,3.7734375,3.875,3.86328125
|
198 |
Service: Live Chat Insights,Summary,Auto,Mixtral 8x7B,Mixtral-8x7B-v0.1,Mistral,3.8046875,3.96875,3.6015625,3.953125,3.83203125
|
@@ -202,10 +202,10 @@ Service: Knowledge creation from Case Info,Generation,Manual,GPT 4 Turbo,gpt-4-0
|
|
202 |
Service: Knowledge creation from Case Info,Generation,Manual,Mixtral 8x7B,Mixtral-8x7B-v0.1,Mistral,3.354166666666667,3.1197916666666665,3.0729166666666665,3.223958333333333,3.192708333333333
|
203 |
Service: Knowledge creation from Case Info,Generation,Manual,AI21 Jamba-Instruct,AI21 (jamba-instruct-preview),AI21,3.1770833333333335,2.9010416666666665,3.0416666666666665,2.901041666666667,3.005208333333333
|
204 |
Service: Knowledge creation from Case Info,Generation,Manual,Mistral 7B,Mistral-7B-Instruct-v0.1,Mistral,3.234375,3.0364583333333335,3.1666666666666665,3.0885416666666665,3.1315104166666665
|
205 |
-
Service: Knowledge creation from Case Info,Generation,Manual,XGen
|
206 |
Service: Knowledge creation from Case Info,Generation,Manual,Claude 3 Haiku,Claude 3 Haiku,Anthropic,2.979166666666667,3.1562499999999996,2.8125000000000004,2.8385416666666665,2.946614583333333
|
207 |
Service: Knowledge creation from Case Info,Generation,Manual,Cohere Command Text,cohere.command-text-v14,Cohere AI,3.03125,2.8177083333333335,3.09375,2.6822916666666665,2.90625
|
208 |
-
Service: Knowledge creation from Case Info,Generation,Manual,Gemini
|
209 |
Service: Knowledge creation from Case Info,Generation,Manual,LLaMA 3 70B,Meta-Llama-3-70B-Instruct,Meta,3.0729166666666665,2.4687500000000004,2.9947916666666665,2.416666666666667,2.73828125
|
210 |
Service: Knowledge creation from Case Info,Generation,Manual,LLaMA 3 8B,Meta-Llama-3-8B-Instruct,Meta,3.0572916666666665,2.770833333333333,3.0520833333333335,2.8489583333333335,2.932291666666667
|
211 |
Service: Knowledge creation from Case Info,Generation,Manual,SF-TextBase 7B,TextBase-7B (Mistral FT),Salesforce,3.1562500000000004,2.927083333333333,3.1197916666666665,3.0989583333333335,3.0755208333333335
|
@@ -217,10 +217,10 @@ Service: Reply Recommendations,Generation,Manual,GPT 4 Turbo,gpt-4-0613,OpenAI,3
|
|
217 |
Service: Reply Recommendations,Generation,Manual,Mixtral 8x7B,Mixtral-8x7B-v0.1,Mistral,3.015740740740741,3.0203703703703706,3.2111111111111112,2.837037037037037,3.021064814814815
|
218 |
Service: Reply Recommendations,Generation,Manual,AI21 Jamba-Instruct,AI21 (jamba-instruct-preview),AI21,3.175,2.733333333333333,2.977777777777778,3.0962962962962965,2.995601851851852
|
219 |
Service: Reply Recommendations,Generation,Manual,Mistral 7B,Mistral-7B-Instruct-v0.1,Mistral,2.5185185185185186,2.4657407407407406,2.660185185185185,2.4314814814814816,2.5189814814814815
|
220 |
-
Service: Reply Recommendations,Generation,Manual,XGen
|
221 |
Service: Reply Recommendations,Generation,Manual,Claude 3 Haiku,Claude 3 Haiku,Anthropic,3.3435185185185192,3.1203703703703702,3.184259259259259,3.28425925925926,3.233101851851852
|
222 |
Service: Reply Recommendations,Generation,Manual,Cohere Command Text,cohere.command-text-v14,Cohere AI,2.3768518518518515,2.8157407407407415,2.813888888888889,2.641666666666667,2.6620370370370376
|
223 |
-
Service: Reply Recommendations,Generation,Manual,Gemini
|
224 |
Service: Reply Recommendations,Generation,Manual,LLaMA 3 70B,Meta-Llama-3-70B-Instruct,Meta,3.2796296296296297,3.2101851851851846,3.230555555555556,3.021296296296297,3.185416666666667
|
225 |
Service: Reply Recommendations,Generation,Manual,LLaMA 3 8B,Meta-Llama-3-8B-Instruct,Meta,3.087037037037037,3.1166666666666667,3.249074074074074,2.888888888888889,3.0854166666666667
|
226 |
Service: Call Summary,Summary,Manual,GPT 3.5 Turbo,gpt-3.5-turbo,OpenAI,3.5694444444444446,3.3541666666666665,3.7948717948717947,3.4193548387096775,3.534459436173146
|
@@ -228,10 +228,10 @@ Service: Call Summary,Summary,Manual,GPT 4 Turbo,gpt-4-0613,OpenAI,3.48051948051
|
|
228 |
Service: Call Summary,Summary,Manual,Mixtral 8x7B,Mixtral-8x7B-v0.1,Mistral,2.721518987341772,2.6962025316455698,2.779220779220779,2.6835443037974684,2.7201216505013974
|
229 |
Service: Call Summary,Summary,Manual,AI21 Jamba-Instruct,AI21 (jamba-instruct-preview),AI21,2.5844155844155843,2.7088607594936707,2.74025974025974,2.6493506493506493,2.6707216833799112
|
230 |
Service: Call Summary,Summary,Manual,Mistral 7B,Mistral-7B-Instruct-v0.1,Mistral,2.7468354430379747,2.7721518987341773,2.8666666666666667,2.6962025316455698,2.770464135021097
|
231 |
-
Service: Call Summary,Summary,Manual,XGen
|
232 |
Service: Call Summary,Summary,Manual,Claude 3 Haiku,Claude 3 Haiku,Anthropic,2.518987341772152,2.5949367088607596,2.7183098591549295,2.5822784810126582,2.603628097700125
|
233 |
Service: Call Summary,Summary,Manual,Cohere Command Text,cohere.command-text-v14,Cohere AI,2.3417721518987342,2.4050632911392404,2.5205479452054793,2.367088607594937,2.4086179989595977
|
234 |
-
Service: Call Summary,Summary,Manual,Gemini
|
235 |
Service: Call Summary,Summary,Manual,LLaMA 3 70B,Meta-Llama-3-70B-Instruct,Meta,2.7435897435897436,2.769230769230769,2.9285714285714284,2.6666666666666665,2.777014652014652
|
236 |
Service: Call Summary,Summary,Manual,LLaMA 3 8B,Meta-Llama-3-8B-Instruct,Meta,2.5135135135135136,2.6052631578947367,2.6818181818181817,2.4871794871794872,2.57194358510148
|
237 |
Service: Live Chat Insights,Summary,Manual,GPT 3.5 Turbo,gpt-3.5-turbo,OpenAI,3.72,3.7,3.7,3.74,3.7150000000000003
|
@@ -239,10 +239,10 @@ Service: Live Chat Insights,Summary,Manual,GPT 4 Turbo,gpt-4-0613,OpenAI,3.72,3.
|
|
239 |
Service: Live Chat Insights,Summary,Manual,Mixtral 8x7B,Mixtral-8x7B-v0.1,Mistral,3.56,3.6,3.1,3.68,3.485
|
240 |
Service: Live Chat Insights,Summary,Manual,AI21 Jamba-Instruct,AI21 (jamba-instruct-preview),AI21,3.56,3.64,3.2,3.68,3.52
|
241 |
Service: Live Chat Insights,Summary,Manual,Mistral 7B,Mistral-7B-Instruct-v0.1,Mistral,3.56,3.54,3.2,3.6,3.475
|
242 |
-
Service: Live Chat Insights,Summary,Manual,XGen
|
243 |
Service: Live Chat Insights,Summary,Manual,Claude 3 Haiku,Claude 3 Haiku,Anthropic,3.64,3.6,2.92,3.64,3.45
|
244 |
Service: Live Chat Insights,Summary,Manual,Cohere Command Text,cohere.command-text-v14,Cohere AI,3.26,3.36,3.0,3.56,3.295
|
245 |
-
Service: Live Chat Insights,Summary,Manual,Gemini
|
246 |
Service: Live Chat Insights,Summary,Manual,LLaMA 3 70B,Meta-Llama-3-70B-Instruct,Meta,3.7,3.64,3.24,3.7,3.5700000000000003
|
247 |
Service: Live Chat Insights,Summary,Manual,LLaMA 3 8B,Meta-Llama-3-8B-Instruct,Meta,3.66,3.62,3.14,3.66,3.52
|
248 |
Service: Live Chat Insights,Summary,Manual,SF-TextSum,Summarization model 7B for Service (Mistral FT),Salesforce,3.46,3.56,3.58,3.56,3.54
|
@@ -254,10 +254,10 @@ Sales: Call Summary,Summary,Manual,GPT 4 Turbo,gpt-4-0613,OpenAI,3.1851851851851
|
|
254 |
Sales: Call Summary,Summary,Manual,Mixtral 8x7B,Mixtral-8x7B-v0.1,Mistral,2.5,2.5370370370370368,2.509259259259259,2.5462962962962963,2.523148148148148
|
255 |
Sales: Call Summary,Summary,Manual,AI21 Jamba-Instruct,AI21 (jamba-instruct-preview),AI21,2.351851851851852,2.5555555555555554,2.7407407407407405,2.3796296296296293,2.506944444444444
|
256 |
Sales: Call Summary,Summary,Manual,Mistral 7B,Mistral-7B-Instruct-v0.1,Mistral,2.4444444444444446,2.6203703703703707,2.7685185185185186,2.3333333333333335,2.541666666666667
|
257 |
-
Sales: Call Summary,Summary,Manual,XGen
|
258 |
Sales: Call Summary,Summary,Manual,Claude 3 Haiku,Claude 3 Haiku,Anthropic,2.898148148148148,2.75,2.935185185185185,2.5555555555555554,2.784722222222222
|
259 |
Sales: Call Summary,Summary,Manual,Cohere Command Text,cohere.command-text-v14,Cohere AI,2.074074074074074,2.2314814814814814,2.3703703703703702,1.9351851851851853,2.1527777777777777
|
260 |
-
Sales: Call Summary,Summary,Manual,Gemini
|
261 |
Sales: Call Summary,Summary,Manual,LLaMA 3 70B,Meta-Llama-3-70B-Instruct,Meta,2.657407407407407,2.6018518518518516,3.046296296296296,2.3055555555555554,2.6527777777777772
|
262 |
Sales: Call Summary,Summary,Manual,LLaMA 3 8B,Meta-Llama-3-8B-Instruct,Meta,2.4814814814814814,2.685185185185185,2.768518518518518,2.2129629629629632,2.5370370370370368
|
263 |
Sales: Call Summary,Summary,Manual,SF-TextBase 7B,TextBase-7B (Mistral FT),Salesforce,2.490740740740741,2.5462962962962963,2.8240740740740744,2.1203703703703702,2.4953703703703702
|
@@ -269,10 +269,10 @@ Sales: Email Generation,Generation,Manual,GPT 4 Turbo,gpt-4-0613,OpenAI,3.361635
|
|
269 |
Sales: Email Generation,Generation,Manual,Mixtral 8x7B,Mixtral-8x7B-v0.1,Mistral,3.2317610062893083,2.4915094339622645,3.0503144654088055,2.949685534591195,2.9308176100628933
|
270 |
Sales: Email Generation,Generation,Manual,AI21 Jamba-Instruct,AI21 (jamba-instruct-preview),AI21,3.162758310871518,2.105031446540881,2.833333333333333,2.986477987421384,2.771900269541779
|
271 |
Sales: Email Generation,Generation,Manual,Mistral 7B,Mistral-7B-Instruct-v0.1,Mistral,3.0338274932614566,1.855031446540881,2.9367924528301894,2.598787061994609,2.606109613656784
|
272 |
-
Sales: Email Generation,Generation,Manual,XGen
|
273 |
Sales: Email Generation,Generation,Manual,Claude 3 Haiku,Claude 3 Haiku,Anthropic,3.111635220125786,2.5229559748427675,3.061006289308176,3.0185534591194974,2.9285377358490567
|
274 |
Sales: Email Generation,Generation,Manual,Cohere Command Text,cohere.command-text-v14,Cohere AI,2.912309074573226,2.0734950584007192,2.843800539083558,2.5796046720575014,2.602302336028751
|
275 |
-
Sales: Email Generation,Generation,Manual,Gemini
|
276 |
Sales: Email Generation,Generation,Manual,LLaMA 3 70B,Meta-Llama-3-70B-Instruct,Meta,3.0899371069182386,2.9817610062893087,3.155974842767295,3.08993710691824,3.0794025157232707
|
277 |
Sales: Email Generation,Generation,Manual,LLaMA 3 8B,Meta-Llama-3-8B-Instruct,Meta,3.251572327044026,3.0471698113207544,3.0764150943396227,3.133647798742139,3.1272012578616355
|
278 |
Sales: Email Generation,Generation,Manual,SF-TextBase 7B,TextBase-7B (Mistral FT),Salesforce,3.155974842767295,2.359119496855346,3.0169811320754722,2.7386792452830186,2.8176886792452827
|
|
|
1 |
Use Case Name,Use Case Type,Accuracy Method,Model Name,Model Version,LLM Provider,Factuality,Instruction Following,Conciseness,Completeness,Accuracy
|
2 |
+
Service: Conversation summary,Summary,Auto,Gemini Pro 1,Gemini Pro 1,Google,3.871720116618076,3.9834791059280854,3.847424684159378,3.9193391642371234,3.9054907677356656
|
3 |
Service: Conversation summary,Summary,Auto,GPT4-o,GPT4-o,OpenAI,3.9669582118561713,3.9961127308066082,3.9300291545189503,3.9844509232264333,3.9693877551020407
|
4 |
Service: Conversation summary,Summary,Auto,AI21 Jamba-Instruct,AI21 (jamba-instruct-preview),AI21,3.9310009718172982,3.998056365403304,3.8104956268221573,3.9737609329446064,3.9283284742468414
|
5 |
Service: Conversation summary,Summary,Auto,GPT 4 Turbo,gpt-4-0613,OpenAI,3.9504373177842567,4.0,3.9067055393586005,3.9805636540330416,3.9594266277939747
|
6 |
Service: Conversation summary,Summary,Auto,SF-TextBase 70B,TextBase-70B (Llama FT),Salesforce,3.9591836734693877,3.998056365403304,3.881438289601555,3.992225461613217,3.957725947521866
|
7 |
Service: Conversation summary,Summary,Auto,Claude 3 Haiku,Claude 3 Haiku,Anthropic,3.9591836734693877,3.999028182701652,3.6997084548104957,3.993197278911565,3.912779397473275
|
8 |
Service: Conversation summary,Summary,Auto,Cohere Command R+,cohere.cmd-R+,Cohere AI,3.954324586977648,4.0,3.8328474246841595,3.9951409135082603,3.945578231292517
|
9 |
+
Service: Conversation summary,Summary,Auto,XGen 2,XGen 2 (1228),Salesforce,3.880466472303207,3.9941690962099123,3.9047619047619047,3.836734693877551,3.9040330417881437
|
10 |
Service: Conversation summary,Summary,Auto,SF-TextBase 7B,TextBase-7B (Mistral FT),Salesforce,3.8746355685131197,3.991253644314869,3.7862001943634596,3.9407191448007777,3.898202137998057
|
11 |
Service: Conversation summary,Summary,Auto,LLaMA 3 70B,Meta-Llama-3-70B-Instruct,Meta,3.9591836734693877,3.999028182701652,3.9280855199222544,3.990281827016521,3.9691448007774537
|
12 |
Service: Conversation summary,Summary,Auto,Cohere Command Text,cohere.command-text-v14,Cohere AI,3.7560738581146746,3.938775510204082,3.6530612244897958,3.938775510204082,3.8216715257531586
|
13 |
Service: Conversation summary,Summary,Auto,GPT 3.5 Turbo,gpt-3.5-turbo,OpenAI,3.9310009718172982,3.9961127308066082,3.9834791059280854,3.935860058309038,3.961613216715257
|
14 |
Service: Conversation summary,Summary,Auto,Claude 3 Opus,Claude 3 (Opus),Anthropic,3.9640427599611274,3.998056365403304,3.7657920310981536,3.989310009718173,3.9293002915451893
|
15 |
+
Service: Conversation summary,Summary,Auto,Gemini Pro 1.5,Gemini Pro 1.5,Google,3.9494655004859087,3.9961127308066082,3.943634596695821,3.9727891156462585,3.9655004859086493
|
16 |
Service: Conversation summary,Summary,Auto,LLaMA 3 8B,Meta-Llama-3-8B-Instruct,Meta,3.922254616132167,4.0,3.565597667638484,3.9961127308066082,3.8709912536443145
|
17 |
Service: Conversation summary,Summary,Auto,Mistral 7B,Mistral-7B-Instruct-v0.1,Mistral,3.7764820213799806,3.927113702623907,3.327502429543246,3.924198250728863,3.738824101068999
|
18 |
Service: Conversation summary,Summary,Auto,Mixtral 8x7B,Mixtral-8x7B-v0.1,Mistral,3.9115646258503403,3.987366375121477,3.3751214771622933,3.9825072886297375,3.814139941690962
|
19 |
Service: Conversation summary,Summary,Auto,SF-TextSum,Summarization model 7B for Service (Mistral FT),Salesforce,3.9037900874635567,3.997084548104956,3.7755102040816326,3.9280855199222544,3.9011175898931
|
20 |
Sales: Email Generation,Generation,Auto,Mistral 7B,Mistral-7B-Instruct-v0.1,Mistral,2.2452830188679247,1.9811320754716981,1.8867924528301887,2.018867924528302,2.0330188679245285
|
21 |
+
Sales: Email Generation,Generation,Auto,Gemini Pro 1,Gemini Pro 1,Google,3.188679245283019,3.207547169811321,3.0754716981132075,3.0377358490566038,3.1273584905660377
|
22 |
Sales: Email Generation,Generation,Auto,LLaMA 3 8B,Meta-Llama-3-8B-Instruct,Meta,3.660377358490566,3.7358490566037736,3.5849056603773586,3.0754716981132075,3.5141509433962264
|
23 |
Sales: Email Generation,Generation,Auto,SF-TextSum,Summarization model 7B for Service (Mistral FT),Salesforce,3.4716981132075473,3.4150943396226414,3.5283018867924527,2.9245283018867925,3.334905660377359
|
24 |
Sales: Email Generation,Generation,Auto,Claude 3 Haiku,Claude 3 Haiku,Anthropic,3.0377358490566038,2.5283018867924527,2.5283018867924527,2.811320754716981,2.7264150943396226
|
25 |
Sales: Email Generation,Generation,Auto,Cohere Command R+,cohere.cmd-R+,Cohere AI,3.7358490566037736,3.6792452830188678,3.547169811320755,3.018867924528302,3.4952830188679247
|
26 |
Sales: Email Generation,Generation,Auto,Cohere Command Text,cohere.command-text-v14,Cohere AI,2.7547169811320753,2.358490566037736,2.2641509433962264,2.452830188679245,2.4575471698113205
|
27 |
+
Sales: Email Generation,Generation,Auto,XGen 2,XGen 2 (1228),Salesforce,2.509433962264151,2.339622641509434,2.3962264150943398,2.2830188679245285,2.3820754716981134
|
28 |
Sales: Email Generation,Generation,Auto,SF-TextBase 70B,TextBase-70B (Llama FT),Salesforce,3.811320754716981,3.849056603773585,3.69811320754717,3.018867924528302,3.5943396226415096
|
29 |
Sales: Email Generation,Generation,Auto,LLaMA 3 70B,Meta-Llama-3-70B-Instruct,Meta,3.811320754716981,3.849056603773585,3.7358490566037736,3.0,3.599056603773585
|
30 |
Sales: Email Generation,Generation,Auto,Claude 3 Opus,Claude 3 (Opus),Anthropic,3.452830188679245,3.056603773584906,3.509433962264151,3.0,3.2547169811320753
|
31 |
Sales: Email Generation,Generation,Auto,GPT4-o,GPT4-o,OpenAI,3.792452830188679,3.830188679245283,3.4716981132075473,3.0377358490566038,3.533018867924528
|
32 |
Sales: Email Generation,Generation,Auto,Mixtral 8x7B,Mixtral-8x7B-v0.1,Mistral,3.0377358490566038,2.9245283018867925,2.7547169811320753,2.69811320754717,2.8537735849056607
|
33 |
+
Sales: Email Generation,Generation,Auto,Gemini Pro 1.5,Gemini Pro 1.5,Google,3.6792452830188678,3.7735849056603774,3.69811320754717,3.0,3.537735849056604
|
34 |
Sales: Email Generation,Generation,Auto,GPT 3.5 Turbo,gpt-3.5-turbo,OpenAI,3.4339622641509435,3.056603773584906,3.056603773584906,2.69811320754717,3.061320754716981
|
35 |
Sales: Email Generation,Generation,Auto,GPT 4 Turbo,gpt-4-0613,OpenAI,3.7358490566037736,3.547169811320755,3.30188679245283,3.0,3.3962264150943398
|
36 |
Sales: Email Generation,Generation,Auto,AI21 Jamba-Instruct,AI21 (jamba-instruct-preview),AI21,2.6226415094339623,2.169811320754717,2.150943396226415,2.490566037735849,2.3584905660377355
|
37 |
Sales: Email Generation,Generation,Auto,SF-TextBase 7B,TextBase-7B (Mistral FT),Salesforce,3.4716981132075473,3.150943396226415,3.150943396226415,2.8867924528301887,3.165094339622642
|
38 |
Service: Reply Recommendations,Generation,Auto,Mistral 7B,Mistral-7B-Instruct-v0.1,Mistral,2.7777777777777777,3.3333333333333335,2.9444444444444446,2.7222222222222223,2.9444444444444446
|
39 |
+
Service: Reply Recommendations,Generation,Auto,Gemini Pro 1,Gemini Pro 1,Google,3.1666666666666665,3.7222222222222223,2.611111111111111,3.5,3.25
|
40 |
Service: Reply Recommendations,Generation,Auto,GPT4-o,GPT4-o,OpenAI,3.388888888888889,3.9444444444444446,3.8333333333333335,3.111111111111111,3.5694444444444446
|
41 |
Service: Reply Recommendations,Generation,Auto,SF-TextBase 70B,TextBase-70B (Llama FT),Salesforce,3.2222222222222223,3.9444444444444446,3.6666666666666665,3.3333333333333335,3.541666666666667
|
42 |
Service: Reply Recommendations,Generation,Auto,LLaMA 3 8B,Meta-Llama-3-8B-Instruct,Meta,3.111111111111111,3.9444444444444446,3.5555555555555554,3.111111111111111,3.4305555555555554
|
|
|
44 |
Service: Reply Recommendations,Generation,Auto,Cohere Command R+,cohere.cmd-R+,Cohere AI,3.0,3.611111111111111,2.888888888888889,3.1666666666666665,3.1666666666666665
|
45 |
Service: Reply Recommendations,Generation,Auto,Cohere Command Text,cohere.command-text-v14,Cohere AI,3.0555555555555554,3.2777777777777777,2.5555555555555554,3.1666666666666665,3.013888888888889
|
46 |
Service: Reply Recommendations,Generation,Auto,SF-TextBase 7B,TextBase-7B (Mistral FT),Salesforce,3.2777777777777777,3.5555555555555554,3.111111111111111,2.8333333333333335,3.1944444444444446
|
47 |
+
Service: Reply Recommendations,Generation,Auto,XGen 2,XGen 2 (1228),Salesforce,3.0,3.7222222222222223,3.2777777777777777,3.111111111111111,3.2777777777777777
|
48 |
Service: Reply Recommendations,Generation,Auto,LLaMA 3 70B,Meta-Llama-3-70B-Instruct,Meta,3.1666666666666665,4.0,3.888888888888889,3.2777777777777777,3.583333333333333
|
49 |
Service: Reply Recommendations,Generation,Auto,Claude 3 Opus,Claude 3 (Opus),Anthropic,3.3333333333333335,4.0,3.388888888888889,3.5555555555555554,3.5694444444444446
|
50 |
Service: Reply Recommendations,Generation,Auto,Mixtral 8x7B,Mixtral-8x7B-v0.1,Mistral,3.2222222222222223,3.888888888888889,3.611111111111111,3.1666666666666665,3.4722222222222223
|
51 |
+
Service: Reply Recommendations,Generation,Auto,Gemini Pro 1.5,Gemini Pro 1.5,Google,3.2777777777777777,3.7777777777777777,3.8333333333333335,2.888888888888889,3.4444444444444446
|
52 |
Service: Reply Recommendations,Generation,Auto,GPT 3.5 Turbo,gpt-3.5-turbo,OpenAI,3.1666666666666665,3.888888888888889,3.7777777777777777,3.0555555555555554,3.472222222222222
|
53 |
Service: Reply Recommendations,Generation,Auto,GPT 4 Turbo,gpt-4-0613,OpenAI,3.2222222222222223,3.9444444444444446,3.5,3.1666666666666665,3.458333333333333
|
54 |
Service: Reply Recommendations,Generation,Auto,AI21 Jamba-Instruct,AI21 (jamba-instruct-preview),AI21,3.0,3.8333333333333335,3.2777777777777777,3.2222222222222223,3.333333333333333
|
55 |
Service: Reply Recommendations,Generation,Auto,SF-TextSum,Summarization model 7B for Service (Mistral FT),Salesforce,3.2777777777777777,3.888888888888889,3.2777777777777777,3.1666666666666665,3.4027777777777777
|
56 |
+
Sales & Service: Update CRM Info,Generation,Auto,Gemini Pro 1,Gemini Pro 1,Google,3.466666666666667,3.4,3.4451219512195124,3.4242424242424243,3.434007760532151
|
57 |
Sales & Service: Update CRM Info,Generation,Auto,GPT4-o,GPT4-o,OpenAI,3.909090909090909,3.8848484848484848,3.915151515151515,3.6484848484848484,3.83939393939394
|
58 |
Sales & Service: Update CRM Info,Generation,Auto,AI21 Jamba-Instruct,AI21 (jamba-instruct-preview),AI21,3.6787878787878787,3.581818181818182,3.709090909090909,3.533333333333333,3.625757575757576
|
59 |
Sales & Service: Update CRM Info,Generation,Auto,GPT 4 Turbo,gpt-4-0613,OpenAI,3.7333333333333334,3.709090909090909,3.806060606060606,3.6363636363636362,3.7212121212121216
|
60 |
Sales & Service: Update CRM Info,Generation,Auto,SF-TextBase 70B,TextBase-70B (Llama FT),Salesforce,3.806060606060606,3.824242424242424,3.83030303030303,3.775757575757576,3.809090909090909
|
61 |
Sales & Service: Update CRM Info,Generation,Auto,Claude 3 Haiku,Claude 3 Haiku,Anthropic,3.6666666666666665,3.618181818181818,3.7151515151515153,3.624242424242424,3.656060606060606
|
62 |
Sales & Service: Update CRM Info,Generation,Auto,Cohere Command R+,cohere.cmd-R+,Cohere AI,3.812121212121212,3.7636363636363637,3.757575757575758,3.624242424242424,3.7393939393939393
|
63 |
+
Sales & Service: Update CRM Info,Generation,Auto,XGen 2,XGen 2 (1228),Salesforce,3.7454545454545456,3.7333333333333334,3.7818181818181817,3.6787878787878787,3.734848484848485
|
64 |
Sales & Service: Update CRM Info,Generation,Auto,SF-TextBase 7B,TextBase-7B (Mistral FT),Salesforce,3.096969696969697,2.9696969696969697,2.8,3.0545454545454547,2.9803030303030305
|
65 |
Sales & Service: Update CRM Info,Generation,Auto,LLaMA 3 70B,Meta-Llama-3-70B-Instruct,Meta,3.812121212121212,3.806060606060606,3.8181818181818183,3.7515151515151515,3.796969696969697
|
66 |
Sales & Service: Update CRM Info,Generation,Auto,Cohere Command Text,cohere.command-text-v14,Cohere AI,2.903030303030303,2.909090909090909,2.8666666666666667,2.896969696969697,2.893939393939394
|
67 |
Sales & Service: Update CRM Info,Generation,Auto,GPT 3.5 Turbo,gpt-3.5-turbo,OpenAI,3.7515151515151515,3.703030303030303,3.7515151515151515,3.6606060606060606,3.716666666666667
|
68 |
Sales & Service: Update CRM Info,Generation,Auto,Claude 3 Opus,Claude 3 (Opus),Anthropic,3.7636363636363637,3.7454545454545456,3.812121212121212,3.606060606060606,3.731818181818182
|
69 |
+
Sales & Service: Update CRM Info,Generation,Auto,Gemini Pro 1.5,Gemini Pro 1.5,Google,3.503030303030303,3.478787878787879,3.5393939393939395,3.3575757575757574,3.4696969696969697
|
70 |
Sales & Service: Update CRM Info,Generation,Auto,LLaMA 3 8B,Meta-Llama-3-8B-Instruct,Meta,3.3575757575757574,2.787878787878788,2.3333333333333335,3.242424242424242,2.93030303030303
|
71 |
Sales & Service: Update CRM Info,Generation,Auto,Mistral 7B,Mistral-7B-Instruct-v0.1,Mistral,1.6787878787878787,1.6606060606060606,1.6424242424242423,1.6848484848484848,1.6666666666666667
|
72 |
Sales & Service: Update CRM Info,Generation,Auto,Mixtral 8x7B,Mixtral-8x7B-v0.1,Mistral,3.690909090909091,3.6666666666666665,3.690909090909091,3.6363636363636362,3.6712121212121214
|
73 |
Sales & Service: Update CRM Info,Generation,Auto,SF-TextSum,Summarization model 7B for Service (Mistral FT),Salesforce,3.1575757575757577,2.981818181818182,3.012121212121212,3.139393939393939,3.0727272727272728
|
74 |
+
Service: Live Chat Summary,Summary,Auto,Gemini Pro 1,Gemini Pro 1,Google,3.381,3.542,3.303,3.479,3.42625
|
75 |
Service: Live Chat Summary,Summary,Auto,GPT4-o,GPT4-o,OpenAI,3.641,3.965,3.946,3.436,3.747
|
76 |
Service: Live Chat Summary,Summary,Auto,AI21 Jamba-Instruct,AI21 (jamba-instruct-preview),AI21,3.6236236236236237,3.715,3.714,3.331,3.595905905905906
|
77 |
Service: Live Chat Summary,Summary,Auto,SF-TextBase 70B,TextBase-70B (Llama FT),Salesforce,3.718,3.882,3.904,3.4004004004004003,3.7261001001001
|
78 |
Service: Live Chat Summary,Summary,Auto,Claude 3 Haiku,Claude 3 Haiku,Anthropic,3.588,3.927,3.607,3.566,3.6720000000000006
|
79 |
Service: Live Chat Summary,Summary,Auto,Cohere Command R+,cohere.cmd-R+,Cohere AI,3.453,3.945,3.666,3.725,3.69725
|
80 |
+
Service: Live Chat Summary,Summary,Auto,XGen 2,XGen 2 (1228),Salesforce,3.605,3.703,3.689,3.247,3.561
|
81 |
Service: Live Chat Summary,Summary,Auto,SF-TextBase 7B,TextBase-7B (Mistral FT),Salesforce,3.593,3.694,3.75,3.158,3.54875
|
82 |
Service: Live Chat Summary,Summary,Auto,LLaMA 3 70B,Meta-Llama-3-70B-Instruct,Meta,3.765,3.903,3.947,3.309,3.731
|
83 |
Service: Live Chat Summary,Summary,Auto,Cohere Command Text,cohere.command-text-v14,Cohere AI,3.502,3.744,3.533,3.497,3.569
|
84 |
Service: Live Chat Summary,Summary,Auto,GPT 4 Turbo,gpt-4-0613,OpenAI,3.587,3.992,3.914,3.822,3.82875
|
85 |
Service: Live Chat Summary,Summary,Auto,GPT 3.5 Turbo,gpt-3.5-turbo,OpenAI,3.717,3.835,3.891,3.164164164164164,3.651791041041041
|
86 |
Service: Live Chat Summary,Summary,Auto,Claude 3 Opus,Claude 3 (Opus),Anthropic,3.574,3.862,3.864,3.492,3.698
|
87 |
+
Service: Live Chat Summary,Summary,Auto,Gemini Pro 1.5,Gemini Pro 1.5,Google,3.396,3.724,3.567,3.635,3.5805
|
88 |
Service: Live Chat Summary,Summary,Auto,LLaMA 3 8B,Meta-Llama-3-8B-Instruct,Meta,3.681,3.93,3.772,3.7,3.7707500000000005
|
89 |
Service: Live Chat Summary,Summary,Auto,Mistral 7B,Mistral-7B-Instruct-v0.1,Mistral,3.492,3.942,3.831,3.353,3.6544999999999996
|
90 |
Service: Live Chat Summary,Summary,Auto,Mixtral 8x7B,Mixtral-8x7B-v0.1,Mistral,3.609,3.866,3.846,3.379,3.6750000000000003
|
91 |
Service: Live Chat Summary,Summary,Auto,SF-TextSum,Summarization model 7B for Service (Mistral FT),Salesforce,3.504,3.595,3.709,3.129129129129129,3.4842822822822823
|
92 |
+
Service: Email Summary,Summary,Auto,Gemini Pro 1,Gemini Pro 1,Google,3.561224489795918,3.7755102040816326,3.4285714285714284,3.7857142857142856,3.6377551020408165
|
93 |
Service: Email Summary,Summary,Auto,GPT4-o,GPT4-o,OpenAI,3.752577319587629,3.9381443298969074,3.8762886597938144,3.9587628865979383,3.8814432989690726
|
94 |
Service: Email Summary,Summary,Auto,AI21 Jamba-Instruct,AI21 (jamba-instruct-preview),AI21,3.377551020408163,3.6122448979591835,3.4285714285714284,3.5306122448979593,3.4872448979591835
|
95 |
Service: Email Summary,Summary,Auto,SF-TextBase 70B,TextBase-70B (Llama FT),Salesforce,3.673469387755102,3.989795918367347,3.9693877551020407,3.979591836734694,3.9030612244897958
|
96 |
Service: Email Summary,Summary,Auto,Claude 3 Haiku,Claude 3 Haiku,Anthropic,3.857142857142857,3.979591836734694,3.673469387755102,3.9591836734693877,3.86734693877551
|
97 |
Service: Email Summary,Summary,Auto,Cohere Command R+,cohere.cmd-R+,Cohere AI,3.693877551020408,4.0,3.857142857142857,4.0,3.8877551020408165
|
98 |
+
Service: Email Summary,Summary,Auto,XGen 2,XGen 2 (1228),Salesforce,3.288659793814433,3.5670103092783507,3.5463917525773194,3.268041237113402,3.4175257731958766
|
99 |
Service: Email Summary,Summary,Auto,SF-TextBase 7B,TextBase-7B (Mistral FT),Salesforce,3.479591836734694,3.7755102040816326,3.5816326530612246,3.8979591836734695,3.683673469387755
|
100 |
Service: Email Summary,Summary,Auto,LLaMA 3 70B,Meta-Llama-3-70B-Instruct,Meta,3.622448979591837,4.0,3.979591836734694,3.989795918367347,3.8979591836734695
|
101 |
Service: Email Summary,Summary,Auto,Cohere Command Text,cohere.command-text-v14,Cohere AI,3.3877551020408165,3.9591836734693877,3.7448979591836733,3.836734693877551,3.732142857142857
|
102 |
Service: Email Summary,Summary,Auto,GPT 4 Turbo,gpt-4-0613,OpenAI,3.795918367346939,3.989795918367347,3.8469387755102042,4.0,3.908163265306123
|
103 |
Service: Email Summary,Summary,Auto,GPT 3.5 Turbo,gpt-3.5-turbo,OpenAI,3.5918367346938775,3.989795918367347,3.9693877551020407,3.86734693877551,3.854591836734694
|
104 |
Service: Email Summary,Summary,Auto,Claude 3 Opus,Claude 3 (Opus),Anthropic,3.795918367346939,4.0,3.8469387755102042,3.979591836734694,3.9056122448979593
|
105 |
+
Service: Email Summary,Summary,Auto,Gemini Pro 1.5,Gemini Pro 1.5,Google,3.5510204081632653,3.989795918367347,3.938775510204082,3.9183673469387754,3.849489795918367
|
106 |
Service: Email Summary,Summary,Auto,LLaMA 3 8B,Meta-Llama-3-8B-Instruct,Meta,3.693877551020408,4.0,3.86734693877551,3.979591836734694,3.885204081632653
|
107 |
Service: Email Summary,Summary,Auto,Mistral 7B,Mistral-7B-Instruct-v0.1,Mistral,3.642857142857143,3.8877551020408165,3.520408163265306,3.8979591836734695,3.737244897959184
|
108 |
Service: Email Summary,Summary,Auto,Mixtral 8x7B,Mixtral-8x7B-v0.1,Mistral,3.673469387755102,3.979591836734694,3.7346938775510203,3.9591836734693877,3.836734693877551
|
109 |
Service: Email Summary,Summary,Auto,SF-TextSum,Summarization model 7B for Service (Mistral FT),Salesforce,3.377551020408163,3.377551020408163,3.36734693877551,3.6530612244897958,3.443877551020408
|
110 |
+
Service: Knowledge creation from Case Info,Generation,Auto,Gemini Pro 1,Gemini Pro 1,Google,3.0,3.1875,2.8125,3.0625,3.015625
|
111 |
Service: Knowledge creation from Case Info,Generation,Auto,GPT4-o,GPT4-o,OpenAI,3.3125,3.625,3.1875,3.6875,3.453125
|
112 |
Service: Knowledge creation from Case Info,Generation,Auto,AI21 Jamba-Instruct,AI21 (jamba-instruct-preview),AI21,3.125,3.125,2.8125,3.125,3.046875
|
113 |
Service: Knowledge creation from Case Info,Generation,Auto,SF-TextBase 70B,TextBase-70B (Llama FT),Salesforce,3.1875,3.5625,3.25,3.4375,3.359375
|
114 |
Service: Knowledge creation from Case Info,Generation,Auto,Claude 3 Haiku,Claude 3 Haiku,Anthropic,3.125,3.5625,3.1875,3.25,3.28125
|
115 |
Service: Knowledge creation from Case Info,Generation,Auto,Cohere Command R+,cohere.cmd-R+,Cohere AI,3.0,3.5625,3.125,3.1875,3.21875
|
116 |
+
Service: Knowledge creation from Case Info,Generation,Auto,XGen 2,XGen 2 (1228),Salesforce,3.1875,3.25,3.0,3.1875,3.15625
|
117 |
Service: Knowledge creation from Case Info,Generation,Auto,SF-TextBase 7B,TextBase-7B (Mistral FT),Salesforce,3.3125,3.5625,3.0625,3.375,3.328125
|
118 |
Service: Knowledge creation from Case Info,Generation,Auto,LLaMA 3 70B,Meta-Llama-3-70B-Instruct,Meta,3.0625,2.4375,2.0625,2.3125,2.46875
|
119 |
Service: Knowledge creation from Case Info,Generation,Auto,Cohere Command Text,cohere.command-text-v14,Cohere AI,3.125,3.0625,2.8125,3.0625,3.015625
|
120 |
Service: Knowledge creation from Case Info,Generation,Auto,GPT 4 Turbo,gpt-4-0613,OpenAI,3.3125,3.6875,3.1875,3.375,3.390625
|
121 |
Service: Knowledge creation from Case Info,Generation,Auto,GPT 3.5 Turbo,gpt-3.5-turbo,OpenAI,3.25,3.5625,3.3125,3.4375,3.390625
|
122 |
Service: Knowledge creation from Case Info,Generation,Auto,Claude 3 Opus,Claude 3 (Opus),Anthropic,3.125,3.375,2.9375,3.1875,3.15625
|
123 |
+
Service: Knowledge creation from Case Info,Generation,Auto,Gemini Pro 1.5,Gemini Pro 1.5,Google,3.1875,3.4375,3.125,3.125,3.21875
|
124 |
Service: Knowledge creation from Case Info,Generation,Auto,LLaMA 3 8B,Meta-Llama-3-8B-Instruct,Meta,3.1875,3.5625,3.0,3.4375,3.296875
|
125 |
Service: Knowledge creation from Case Info,Generation,Auto,Mistral 7B,Mistral-7B-Instruct-v0.1,Mistral,3.0625,3.5625,2.9375,3.375,3.234375
|
126 |
Service: Knowledge creation from Case Info,Generation,Auto,Mixtral 8x7B,Mixtral-8x7B-v0.1,Mistral,3.25,3.4375,3.125,3.375,3.296875
|
127 |
Service: Knowledge creation from Case Info,Generation,Auto,SF-TextSum,Summarization model 7B for Service (Mistral FT),Salesforce,3.125,3.4375,3.0,3.3125,3.21875
|
128 |
+
Sales: Email Summary,Summary,Auto,Gemini Pro 1,Gemini Pro 1,Google,3.377551020408163,3.3877551020408165,3.193877551020408,3.4285714285714284,3.346938775510204
|
129 |
Sales: Email Summary,Summary,Auto,GPT4-o,GPT4-o,OpenAI,3.8877551020408165,3.9693877551020407,3.86734693877551,3.8979591836734695,3.9056122448979593
|
130 |
Sales: Email Summary,Summary,Auto,AI21 Jamba-Instruct,AI21 (jamba-instruct-preview),AI21,3.5714285714285716,3.8979591836734695,3.9183673469387754,3.663265306122449,3.7627551020408165
|
131 |
Sales: Email Summary,Summary,Auto,SF-TextBase 70B,TextBase-70B (Llama FT),Salesforce,3.663265306122449,3.979591836734694,3.9591836734693877,3.836734693877551,3.8596938775510203
|
132 |
Sales: Email Summary,Summary,Auto,Claude 3 Haiku,Claude 3 Haiku,Anthropic,3.857142857142857,3.989795918367347,3.9183673469387754,3.7653061224489797,3.88265306122449
|
133 |
Sales: Email Summary,Summary,Auto,Cohere Command R+,cohere.cmd-R+,Cohere AI,3.7244897959183674,3.8979591836734695,3.795918367346939,3.7551020408163267,3.793367346938776
|
134 |
+
Sales: Email Summary,Summary,Auto,XGen 2,XGen 2 (1228),Salesforce,3.2783505154639174,2.979381443298969,3.142857142857143,2.8969072164948453,3.0743740795287184
|
135 |
Sales: Email Summary,Summary,Auto,SF-TextBase 7B,TextBase-7B (Mistral FT),Salesforce,3.357142857142857,3.6020408163265305,3.663265306122449,3.510204081632653,3.5331632653061225
|
136 |
Sales: Email Summary,Summary,Auto,LLaMA 3 70B,Meta-Llama-3-70B-Instruct,Meta,3.642857142857143,3.9693877551020407,3.9591836734693877,3.8979591836734695,3.86734693877551
|
137 |
Sales: Email Summary,Summary,Auto,Cohere Command Text,cohere.command-text-v14,Cohere AI,3.154639175257732,2.9690721649484537,2.649484536082474,3.2783505154639174,3.0128865979381443
|
138 |
Sales: Email Summary,Summary,Auto,GPT 4 Turbo,gpt-4-0613,OpenAI,3.826530612244898,4.0,3.9285714285714284,3.9591836734693877,3.928571428571429
|
139 |
Sales: Email Summary,Summary,Auto,GPT 3.5 Turbo,gpt-3.5-turbo,OpenAI,3.7857142857142856,3.9591836734693877,3.9693877551020407,3.693877551020408,3.8520408163265305
|
140 |
Sales: Email Summary,Summary,Auto,Claude 3 Opus,Claude 3 (Opus),Anthropic,3.826530612244898,4.0,3.9183673469387754,3.938775510204082,3.920918367346939
|
141 |
+
Sales: Email Summary,Summary,Auto,Gemini Pro 1.5,Gemini Pro 1.5,Google,3.7653061224489797,3.9285714285714284,3.9183673469387754,3.663265306122449,3.8188775510204085
|
142 |
Sales: Email Summary,Summary,Auto,LLaMA 3 8B,Meta-Llama-3-8B-Instruct,Meta,3.7346938775510203,4.0,3.9693877551020407,3.806122448979592,3.877551020408163
|
143 |
Sales: Email Summary,Summary,Auto,Mistral 7B,Mistral-7B-Instruct-v0.1,Mistral,3.6288659793814433,3.948453608247423,3.7653061224489797,3.948453608247423,3.822769829581317
|
144 |
Sales: Email Summary,Summary,Auto,Mixtral 8x7B,Mixtral-8x7B-v0.1,Mistral,3.5306122448979593,3.9591836734693877,3.8469387755102042,3.816326530612245,3.788265306122449
|
145 |
Sales: Email Summary,Summary,Auto,SF-TextSum,Summarization model 7B for Service (Mistral FT),Salesforce,3.5306122448979593,3.836734693877551,3.7346938775510203,3.642857142857143,3.686224489795918
|
146 |
Service: Call Summary,Summary,Auto,Mistral 7B,Mistral-7B-Instruct-v0.1,Mistral,3.75,3.95,3.925,3.925,3.8875
|
147 |
+
Service: Call Summary,Summary,Auto,Gemini Pro 1,Gemini Pro 1,Google,3.775,3.95,3.9,3.85,3.8687500000000004
|
148 |
Service: Call Summary,Summary,Auto,GPT4-o,GPT4-o,OpenAI,3.925,3.95,3.95,3.925,3.9375
|
149 |
Service: Call Summary,Summary,Auto,SF-TextBase 70B,TextBase-70B (Llama FT),Salesforce,3.975,3.95,3.95,3.95,3.9562500000000003
|
150 |
Service: Call Summary,Summary,Auto,LLaMA 3 8B,Meta-Llama-3-8B-Instruct,Meta,3.9,3.95,3.95,3.95,3.9375000000000004
|
|
|
152 |
Service: Call Summary,Summary,Auto,Cohere Command R+,cohere.cmd-R+,Cohere AI,3.925,3.95,3.95,3.925,3.9375
|
153 |
Service: Call Summary,Summary,Auto,Cohere Command Text,cohere.command-text-v14,Cohere AI,3.6,3.75,3.675,3.85,3.7187499999999996
|
154 |
Service: Call Summary,Summary,Auto,SF-TextBase 7B,TextBase-7B (Mistral FT),Salesforce,3.95,4.0,3.975,3.875,3.95
|
155 |
+
Service: Call Summary,Summary,Auto,XGen 2,XGen 2 (1228),Salesforce,3.65,3.95,3.95,3.8,3.8375
|
156 |
Service: Call Summary,Summary,Auto,LLaMA 3 70B,Meta-Llama-3-70B-Instruct,Meta,3.875,3.95,3.95,3.925,3.925
|
157 |
Service: Call Summary,Summary,Auto,Claude 3 Opus,Claude 3 (Opus),Anthropic,3.95,4.0,4.0,4.0,3.9875
|
158 |
Service: Call Summary,Summary,Auto,Mixtral 8x7B,Mixtral-8x7B-v0.1,Mistral,3.825,3.975,3.95,3.925,3.91875
|
159 |
+
Service: Call Summary,Summary,Auto,Gemini Pro 1.5,Gemini Pro 1.5,Google,3.95,3.95,4.0,3.925,3.95625
|
160 |
Service: Call Summary,Summary,Auto,GPT 3.5 Turbo,gpt-3.5-turbo,OpenAI,3.85,3.95,3.95,3.875,3.90625
|
161 |
Service: Call Summary,Summary,Auto,GPT 4 Turbo,gpt-4-0613,OpenAI,3.9,3.95,3.95,3.925,3.93125
|
162 |
Service: Call Summary,Summary,Auto,AI21 Jamba-Instruct,AI21 (jamba-instruct-preview),AI21,3.825,3.95,3.925,3.9,3.8999999999999995
|
163 |
Service: Call Summary,Summary,Auto,SF-TextSum,Summarization model 7B for Service (Mistral FT),Salesforce,3.875,3.95,3.95,3.875,3.9125
|
164 |
+
Sales: Call Summary,Summary,Auto,Gemini Pro 1,Gemini Pro 1,Google,2.789473684210526,3.0526315789473686,2.473684210526316,2.8947368421052633,2.8026315789473686
|
165 |
Sales: Call Summary,Summary,Auto,GPT4-o,GPT4-o,OpenAI,3.9473684210526314,3.9473684210526314,3.473684210526316,3.210526315789474,3.6447368421052633
|
166 |
Sales: Call Summary,Summary,Auto,AI21 Jamba-Instruct,AI21 (jamba-instruct-preview),AI21,3.3684210526315788,3.3157894736842106,3.1052631578947367,3.0,3.1973684210526314
|
167 |
Sales: Call Summary,Summary,Auto,SF-TextBase 70B,TextBase-70B (Llama FT),Salesforce,3.6842105263157894,3.3684210526315788,3.263157894736842,2.789473684210526,3.276315789473684
|
168 |
Sales: Call Summary,Summary,Auto,Claude 3 Haiku,Claude 3 Haiku,Anthropic,3.6315789473684212,3.473684210526316,3.3157894736842106,3.0526315789473686,3.368421052631579
|
169 |
Sales: Call Summary,Summary,Auto,Cohere Command R+,cohere.cmd-R+,Cohere AI,3.789473684210526,3.6842105263157894,2.8421052631578947,3.0526315789473686,3.3421052631578942
|
170 |
+
Sales: Call Summary,Summary,Auto,XGen 2,XGen 2 (1228),Salesforce,2.789473684210526,2.736842105263158,2.5789473684210527,2.6315789473684212,2.6842105263157894
|
171 |
Sales: Call Summary,Summary,Auto,SF-TextBase 7B,TextBase-7B (Mistral FT),Salesforce,3.210526315789474,3.1052631578947367,3.210526315789474,2.6315789473684212,3.0394736842105265
|
172 |
Sales: Call Summary,Summary,Auto,LLaMA 3 70B,Meta-Llama-3-70B-Instruct,Meta,3.4210526315789473,3.263157894736842,3.473684210526316,2.8421052631578947,3.2499999999999996
|
173 |
Sales: Call Summary,Summary,Auto,Cohere Command Text,cohere.command-text-v14,Cohere AI,1.7894736842105263,2.473684210526316,2.1052631578947367,2.3157894736842106,2.1710526315789473
|
174 |
Sales: Call Summary,Summary,Auto,GPT 4 Turbo,gpt-4-0613,OpenAI,4.0,3.736842105263158,3.3157894736842106,3.0,3.513157894736842
|
175 |
Sales: Call Summary,Summary,Auto,GPT 3.5 Turbo,gpt-3.5-turbo,OpenAI,3.8421052631578947,3.3684210526315788,3.8421052631578947,2.888888888888889,3.485380116959064
|
176 |
Sales: Call Summary,Summary,Auto,Claude 3 Opus,Claude 3 (Opus),Anthropic,3.9473684210526314,3.8947368421052633,3.5789473684210527,3.0,3.605263157894737
|
177 |
+
Sales: Call Summary,Summary,Auto,Gemini Pro 1.5,Gemini Pro 1.5,Google,3.736842105263158,3.6315789473684212,3.0526315789473686,3.0526315789473686,3.3684210526315788
|
178 |
Sales: Call Summary,Summary,Auto,LLaMA 3 8B,Meta-Llama-3-8B-Instruct,Meta,2.9473684210526314,3.1578947368421053,2.6315789473684212,2.8947368421052633,2.9078947368421053
|
179 |
Sales: Call Summary,Summary,Auto,Mistral 7B,Mistral-7B-Instruct-v0.1,Mistral,3.5555555555555554,3.4210526315789473,3.210526315789474,3.1052631578947367,3.323099415204678
|
180 |
Sales: Call Summary,Summary,Auto,Mixtral 8x7B,Mixtral-8x7B-v0.1,Mistral,3.263157894736842,3.263157894736842,2.6315789473684212,3.0,3.0394736842105265
|
181 |
Sales: Call Summary,Summary,Auto,SF-TextSum,Summarization model 7B for Service (Mistral FT),Salesforce,3.0526315789473686,2.9473684210526314,3.0,2.736842105263158,2.9342105263157894
|
182 |
+
Service: Live Chat Insights,Summary,Auto,Gemini Pro 1,Gemini Pro 1,Google,3.76953125,3.97265625,3.87109375,3.8359375,3.8623046875
|
183 |
Service: Live Chat Insights,Summary,Auto,GPT4-o,GPT4-o,OpenAI,3.91015625,4.0,3.9765625,3.91796875,3.951171875
|
184 |
Service: Live Chat Insights,Summary,Auto,AI21 Jamba-Instruct,AI21 (jamba-instruct-preview),AI21,3.859375,3.97265625,3.8515625,3.8203125,3.8759765625
|
185 |
Service: Live Chat Insights,Summary,Auto,SF-TextBase 70B,TextBase-70B (Llama FT),Salesforce,3.85546875,3.99609375,3.94140625,3.96875,3.9404296875
|
186 |
Service: Live Chat Insights,Summary,Auto,Claude 3 Haiku,Claude 3 Haiku,Anthropic,3.8828125,4.0,3.796875,3.98046875,3.9150390625
|
187 |
Service: Live Chat Insights,Summary,Auto,Cohere Command R+,cohere.cmd-R+,Cohere AI,3.87109375,3.984375,3.8671875,3.9375,3.9150390625
|
188 |
+
Service: Live Chat Insights,Summary,Auto,XGen 2,XGen 2 (1228),Salesforce,3.74609375,3.9765625,3.921875,3.4375,3.7705078125
|
189 |
Service: Live Chat Insights,Summary,Auto,SF-TextBase 7B,TextBase-7B (Mistral FT),Salesforce,3.71484375,3.97265625,3.90625,3.6875,3.8203125
|
190 |
Service: Live Chat Insights,Summary,Auto,LLaMA 3 70B,Meta-Llama-3-70B-Instruct,Meta,3.85546875,3.9921875,3.97265625,3.93359375,3.9384765625
|
191 |
Service: Live Chat Insights,Summary,Auto,Cohere Command Text,cohere.command-text-v14,Cohere AI,3.51171875,3.9375,3.484375,3.8671875,3.7001953125
|
192 |
Service: Live Chat Insights,Summary,Auto,GPT 4 Turbo,gpt-4-0613,OpenAI,3.91015625,3.9921875,3.93359375,3.9375,3.943359375
|
193 |
Service: Live Chat Insights,Summary,Auto,GPT 3.5 Turbo,gpt-3.5-turbo,OpenAI,3.90234375,3.96875,4.0,3.71875,3.8974609375
|
194 |
Service: Live Chat Insights,Summary,Auto,Claude 3 Opus,Claude 3 (Opus),Anthropic,3.9140625,4.0,3.82421875,3.98046875,3.9296875
|
195 |
+
Service: Live Chat Insights,Summary,Auto,Gemini Pro 1.5,Gemini Pro 1.5,Google,3.77734375,3.98828125,3.9375,3.88671875,3.8974609375
|
196 |
Service: Live Chat Insights,Summary,Auto,LLaMA 3 8B,Meta-Llama-3-8B-Instruct,Meta,3.8359375,3.9921875,3.62890625,3.96484375,3.85546875
|
197 |
Service: Live Chat Insights,Summary,Auto,Mistral 7B,Mistral-7B-Instruct-v0.1,Mistral,3.8125,3.9921875,3.7734375,3.875,3.86328125
|
198 |
Service: Live Chat Insights,Summary,Auto,Mixtral 8x7B,Mixtral-8x7B-v0.1,Mistral,3.8046875,3.96875,3.6015625,3.953125,3.83203125
|
|
|
202 |
Service: Knowledge creation from Case Info,Generation,Manual,Mixtral 8x7B,Mixtral-8x7B-v0.1,Mistral,3.354166666666667,3.1197916666666665,3.0729166666666665,3.223958333333333,3.192708333333333
|
203 |
Service: Knowledge creation from Case Info,Generation,Manual,AI21 Jamba-Instruct,AI21 (jamba-instruct-preview),AI21,3.1770833333333335,2.9010416666666665,3.0416666666666665,2.901041666666667,3.005208333333333
|
204 |
Service: Knowledge creation from Case Info,Generation,Manual,Mistral 7B,Mistral-7B-Instruct-v0.1,Mistral,3.234375,3.0364583333333335,3.1666666666666665,3.0885416666666665,3.1315104166666665
|
205 |
+
Service: Knowledge creation from Case Info,Generation,Manual,XGen 2,XGen 2 (1228),Salesforce,2.880208333333333,2.921875,3.0416666666666665,2.84375,2.921875
|
206 |
Service: Knowledge creation from Case Info,Generation,Manual,Claude 3 Haiku,Claude 3 Haiku,Anthropic,2.979166666666667,3.1562499999999996,2.8125000000000004,2.8385416666666665,2.946614583333333
|
207 |
Service: Knowledge creation from Case Info,Generation,Manual,Cohere Command Text,cohere.command-text-v14,Cohere AI,3.03125,2.8177083333333335,3.09375,2.6822916666666665,2.90625
|
208 |
+
Service: Knowledge creation from Case Info,Generation,Manual,Gemini Pro 1,Gemini Pro 1,Google,3.213541666666667,2.869791666666667,3.1093750000000004,2.9427083333333335,3.033854166666667
|
209 |
Service: Knowledge creation from Case Info,Generation,Manual,LLaMA 3 70B,Meta-Llama-3-70B-Instruct,Meta,3.0729166666666665,2.4687500000000004,2.9947916666666665,2.416666666666667,2.73828125
|
210 |
Service: Knowledge creation from Case Info,Generation,Manual,LLaMA 3 8B,Meta-Llama-3-8B-Instruct,Meta,3.0572916666666665,2.770833333333333,3.0520833333333335,2.8489583333333335,2.932291666666667
|
211 |
Service: Knowledge creation from Case Info,Generation,Manual,SF-TextBase 7B,TextBase-7B (Mistral FT),Salesforce,3.1562500000000004,2.927083333333333,3.1197916666666665,3.0989583333333335,3.0755208333333335
|
|
|
217 |
Service: Reply Recommendations,Generation,Manual,Mixtral 8x7B,Mixtral-8x7B-v0.1,Mistral,3.015740740740741,3.0203703703703706,3.2111111111111112,2.837037037037037,3.021064814814815
|
218 |
Service: Reply Recommendations,Generation,Manual,AI21 Jamba-Instruct,AI21 (jamba-instruct-preview),AI21,3.175,2.733333333333333,2.977777777777778,3.0962962962962965,2.995601851851852
|
219 |
Service: Reply Recommendations,Generation,Manual,Mistral 7B,Mistral-7B-Instruct-v0.1,Mistral,2.5185185185185186,2.4657407407407406,2.660185185185185,2.4314814814814816,2.5189814814814815
|
220 |
+
Service: Reply Recommendations,Generation,Manual,XGen 2,XGen 2 (1228),Salesforce,2.892592592592593,2.9074074074074074,3.1231481481481485,2.486111111111111,2.852314814814815
|
221 |
Service: Reply Recommendations,Generation,Manual,Claude 3 Haiku,Claude 3 Haiku,Anthropic,3.3435185185185192,3.1203703703703702,3.184259259259259,3.28425925925926,3.233101851851852
|
222 |
Service: Reply Recommendations,Generation,Manual,Cohere Command Text,cohere.command-text-v14,Cohere AI,2.3768518518518515,2.8157407407407415,2.813888888888889,2.641666666666667,2.6620370370370376
|
223 |
+
Service: Reply Recommendations,Generation,Manual,Gemini Pro 1,Gemini Pro 1,Google,2.851851851851852,2.761111111111111,2.2962962962962963,3.2379629629629627,2.7868055555555555
|
224 |
Service: Reply Recommendations,Generation,Manual,LLaMA 3 70B,Meta-Llama-3-70B-Instruct,Meta,3.2796296296296297,3.2101851851851846,3.230555555555556,3.021296296296297,3.185416666666667
|
225 |
Service: Reply Recommendations,Generation,Manual,LLaMA 3 8B,Meta-Llama-3-8B-Instruct,Meta,3.087037037037037,3.1166666666666667,3.249074074074074,2.888888888888889,3.0854166666666667
|
226 |
Service: Call Summary,Summary,Manual,GPT 3.5 Turbo,gpt-3.5-turbo,OpenAI,3.5694444444444446,3.3541666666666665,3.7948717948717947,3.4193548387096775,3.534459436173146
|
|
|
228 |
Service: Call Summary,Summary,Manual,Mixtral 8x7B,Mixtral-8x7B-v0.1,Mistral,2.721518987341772,2.6962025316455698,2.779220779220779,2.6835443037974684,2.7201216505013974
|
229 |
Service: Call Summary,Summary,Manual,AI21 Jamba-Instruct,AI21 (jamba-instruct-preview),AI21,2.5844155844155843,2.7088607594936707,2.74025974025974,2.6493506493506493,2.6707216833799112
|
230 |
Service: Call Summary,Summary,Manual,Mistral 7B,Mistral-7B-Instruct-v0.1,Mistral,2.7468354430379747,2.7721518987341773,2.8666666666666667,2.6962025316455698,2.770464135021097
|
231 |
+
Service: Call Summary,Summary,Manual,XGen 2,XGen 2 (1228),Salesforce,2.6962025316455698,2.810126582278481,2.961038961038961,2.6962025316455698,2.7908926516521455
|
232 |
Service: Call Summary,Summary,Manual,Claude 3 Haiku,Claude 3 Haiku,Anthropic,2.518987341772152,2.5949367088607596,2.7183098591549295,2.5822784810126582,2.603628097700125
|
233 |
Service: Call Summary,Summary,Manual,Cohere Command Text,cohere.command-text-v14,Cohere AI,2.3417721518987342,2.4050632911392404,2.5205479452054793,2.367088607594937,2.4086179989595977
|
234 |
+
Service: Call Summary,Summary,Manual,Gemini Pro 1,Gemini Pro 1,Google,2.2051282051282053,2.3684210526315788,2.442857142857143,2.217948717948718,2.3085887796414113
|
235 |
Service: Call Summary,Summary,Manual,LLaMA 3 70B,Meta-Llama-3-70B-Instruct,Meta,2.7435897435897436,2.769230769230769,2.9285714285714284,2.6666666666666665,2.777014652014652
|
236 |
Service: Call Summary,Summary,Manual,LLaMA 3 8B,Meta-Llama-3-8B-Instruct,Meta,2.5135135135135136,2.6052631578947367,2.6818181818181817,2.4871794871794872,2.57194358510148
|
237 |
Service: Live Chat Insights,Summary,Manual,GPT 3.5 Turbo,gpt-3.5-turbo,OpenAI,3.72,3.7,3.7,3.74,3.7150000000000003
|
|
|
239 |
Service: Live Chat Insights,Summary,Manual,Mixtral 8x7B,Mixtral-8x7B-v0.1,Mistral,3.56,3.6,3.1,3.68,3.485
|
240 |
Service: Live Chat Insights,Summary,Manual,AI21 Jamba-Instruct,AI21 (jamba-instruct-preview),AI21,3.56,3.64,3.2,3.68,3.52
|
241 |
Service: Live Chat Insights,Summary,Manual,Mistral 7B,Mistral-7B-Instruct-v0.1,Mistral,3.56,3.54,3.2,3.6,3.475
|
242 |
+
Service: Live Chat Insights,Summary,Manual,XGen 2,XGen 2 (1228),Salesforce,3.54,3.6,3.52,3.64,3.575
|
243 |
Service: Live Chat Insights,Summary,Manual,Claude 3 Haiku,Claude 3 Haiku,Anthropic,3.64,3.6,2.92,3.64,3.45
|
244 |
Service: Live Chat Insights,Summary,Manual,Cohere Command Text,cohere.command-text-v14,Cohere AI,3.26,3.36,3.0,3.56,3.295
|
245 |
+
Service: Live Chat Insights,Summary,Manual,Gemini Pro 1,Gemini Pro 1,Google,3.42,3.5,3.46,3.58,3.4899999999999998
|
246 |
Service: Live Chat Insights,Summary,Manual,LLaMA 3 70B,Meta-Llama-3-70B-Instruct,Meta,3.7,3.64,3.24,3.7,3.5700000000000003
|
247 |
Service: Live Chat Insights,Summary,Manual,LLaMA 3 8B,Meta-Llama-3-8B-Instruct,Meta,3.66,3.62,3.14,3.66,3.52
|
248 |
Service: Live Chat Insights,Summary,Manual,SF-TextSum,Summarization model 7B for Service (Mistral FT),Salesforce,3.46,3.56,3.58,3.56,3.54
|
|
|
254 |
Sales: Call Summary,Summary,Manual,Mixtral 8x7B,Mixtral-8x7B-v0.1,Mistral,2.5,2.5370370370370368,2.509259259259259,2.5462962962962963,2.523148148148148
|
255 |
Sales: Call Summary,Summary,Manual,AI21 Jamba-Instruct,AI21 (jamba-instruct-preview),AI21,2.351851851851852,2.5555555555555554,2.7407407407407405,2.3796296296296293,2.506944444444444
|
256 |
Sales: Call Summary,Summary,Manual,Mistral 7B,Mistral-7B-Instruct-v0.1,Mistral,2.4444444444444446,2.6203703703703707,2.7685185185185186,2.3333333333333335,2.541666666666667
|
257 |
+
Sales: Call Summary,Summary,Manual,XGen 2,XGen 2 (1228),Salesforce,2.4074074074074074,2.490740740740741,2.6574074074074074,2.240740740740741,2.4490740740740744
|
258 |
Sales: Call Summary,Summary,Manual,Claude 3 Haiku,Claude 3 Haiku,Anthropic,2.898148148148148,2.75,2.935185185185185,2.5555555555555554,2.784722222222222
|
259 |
Sales: Call Summary,Summary,Manual,Cohere Command Text,cohere.command-text-v14,Cohere AI,2.074074074074074,2.2314814814814814,2.3703703703703702,1.9351851851851853,2.1527777777777777
|
260 |
+
Sales: Call Summary,Summary,Manual,Gemini Pro 1,Gemini Pro 1,Google,2.7777777777777777,2.6666666666666665,2.7870370370370368,2.8333333333333335,2.7662037037037037
|
261 |
Sales: Call Summary,Summary,Manual,LLaMA 3 70B,Meta-Llama-3-70B-Instruct,Meta,2.657407407407407,2.6018518518518516,3.046296296296296,2.3055555555555554,2.6527777777777772
|
262 |
Sales: Call Summary,Summary,Manual,LLaMA 3 8B,Meta-Llama-3-8B-Instruct,Meta,2.4814814814814814,2.685185185185185,2.768518518518518,2.2129629629629632,2.5370370370370368
|
263 |
Sales: Call Summary,Summary,Manual,SF-TextBase 7B,TextBase-7B (Mistral FT),Salesforce,2.490740740740741,2.5462962962962963,2.8240740740740744,2.1203703703703702,2.4953703703703702
|
|
|
269 |
Sales: Email Generation,Generation,Manual,Mixtral 8x7B,Mixtral-8x7B-v0.1,Mistral,3.2317610062893083,2.4915094339622645,3.0503144654088055,2.949685534591195,2.9308176100628933
|
270 |
Sales: Email Generation,Generation,Manual,AI21 Jamba-Instruct,AI21 (jamba-instruct-preview),AI21,3.162758310871518,2.105031446540881,2.833333333333333,2.986477987421384,2.771900269541779
|
271 |
Sales: Email Generation,Generation,Manual,Mistral 7B,Mistral-7B-Instruct-v0.1,Mistral,3.0338274932614566,1.855031446540881,2.9367924528301894,2.598787061994609,2.606109613656784
|
272 |
+
Sales: Email Generation,Generation,Manual,XGen 2,XGen 2 (1228),Salesforce,2.982704402515723,2.037106918238994,2.9817610062893083,2.619811320754717,2.6553459119496856
|
273 |
Sales: Email Generation,Generation,Manual,Claude 3 Haiku,Claude 3 Haiku,Anthropic,3.111635220125786,2.5229559748427675,3.061006289308176,3.0185534591194974,2.9285377358490567
|
274 |
Sales: Email Generation,Generation,Manual,Cohere Command Text,cohere.command-text-v14,Cohere AI,2.912309074573226,2.0734950584007192,2.843800539083558,2.5796046720575014,2.602302336028751
|
275 |
+
Sales: Email Generation,Generation,Manual,Gemini Pro 1,Gemini Pro 1,Google,3.1317610062893073,2.430503144654088,2.7952830188679245,2.947798742138365,2.826336477987421
|
276 |
Sales: Email Generation,Generation,Manual,LLaMA 3 70B,Meta-Llama-3-70B-Instruct,Meta,3.0899371069182386,2.9817610062893087,3.155974842767295,3.08993710691824,3.0794025157232707
|
277 |
Sales: Email Generation,Generation,Manual,LLaMA 3 8B,Meta-Llama-3-8B-Instruct,Meta,3.251572327044026,3.0471698113207544,3.0764150943396227,3.133647798742139,3.1272012578616355
|
278 |
Sales: Email Generation,Generation,Manual,SF-TextBase 7B,TextBase-7B (Mistral FT),Salesforce,3.155974842767295,2.359119496855346,3.0169811320754722,2.7386792452830186,2.8176886792452827
|
crm-results/hf_leaderboard_crm_bias.csv
ADDED
@@ -0,0 +1,19 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
Model Name,CRM Bias
|
2 |
+
LLaMA 3 70B,"98.3% [98.2%, 98.5%]"
|
3 |
+
SF-TextBase 70B,"98.2% [98.0%, 98.4%]"
|
4 |
+
Claude 3 Opus,"97.8% [97.4%, 98.1%]"
|
5 |
+
GPT 4 Turbo,"97.6% [97.3%, 98.0%]"
|
6 |
+
GPT4-o,"97.1% [96.8%, 97.4%]"
|
7 |
+
XGen 2,"97.0% [96.7%, 97.2%]"
|
8 |
+
Gemini Pro 1,"96.9% [96.7%, 97.1%]"
|
9 |
+
Gemini Pro 1.5,"96.8% [96.3%, 97.2%]"
|
10 |
+
GPT 3.5 Turbo,"96.6% [96.1%, 97.0%]"
|
11 |
+
Claude 3 Haiku,"96.5% [96.1%, 96.9%]"
|
12 |
+
Mistral 7B,"96.2% [96.1%, 96.3%]"
|
13 |
+
Cohere Command R+,"96.1% [95.7%, 96.6%]"
|
14 |
+
AI21 Jamba-Instruct,"95.5% [95.1%, 95.9%]"
|
15 |
+
Cohere Command Text,"95.2% [95.0%, 95.3%]"
|
16 |
+
LLaMA 3 8B,"95.1% [94.8%, 95.5%]"
|
17 |
+
Mixtral 8x7B,"94.9% [94.6%, 95.1%]"
|
18 |
+
SF-TextBase 7B,"94.6% [94.1%, 95.1%]"
|
19 |
+
SF-TextSum,"93.9% [93.3%, 94.4%]"
|
crm-results/hf_leaderboard_latency_cost.csv
CHANGED
@@ -33,5 +33,5 @@ SF-TextBase 70B,Long,TextBase-70B-8K,Self-host (p4de.24xlarge),6.52,253.7,28.17,
|
|
33 |
SF-TextBase 70B,Short,TextBase-70B-8K,Self-host (p4de.24xlarge),6.24,249.7,26.96,High,,,,,,,,,,,,,
|
34 |
SF-TextSum,Long,CRM-TSUM-7b-22k-g5 (endpoint),Self-host (g5.48xlarge),8.85,244.0,16.55,High,,,,,,,,,,,,,
|
35 |
SF-TextSum,Short,CRM-TSUM-7b-22k-g5 (endpoint),Self-host (g5.48xlarge),8.34,250.4,15.60,High,,,,,,,,,,,,,
|
36 |
-
XGen
|
37 |
-
XGen
|
|
|
33 |
SF-TextBase 70B,Short,TextBase-70B-8K,Self-host (p4de.24xlarge),6.24,249.7,26.96,High,,,,,,,,,,,,,
|
34 |
SF-TextSum,Long,CRM-TSUM-7b-22k-g5 (endpoint),Self-host (g5.48xlarge),8.85,244.0,16.55,High,,,,,,,,,,,,,
|
35 |
SF-TextSum,Short,CRM-TSUM-7b-22k-g5 (endpoint),Self-host (g5.48xlarge),8.34,250.4,15.60,High,,,,,,,,,,,,,
|
36 |
+
XGen 2,Long,EinsteinXgen2E4DSStreaming (endpoint),Self-host (p4de.24xlarge),3.71,250.0,16.03,High,not able to get response for large token requests (5K-token input),,,,,,,,,,,,
|
37 |
+
XGen 2,Short,EinsteinXgen2E4DSStreaming (endpoint),Self-host (p4de.24xlarge),2.64,250.0,11.40,High,,,,,,,,,,,,,
|
crm-results/hf_leaderboard_ts.csv
CHANGED
@@ -5,15 +5,15 @@ GPT 3.5 Turbo,45%,59%,100%,13%,36%,2%,-,-,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,
|
|
5 |
AI21 Jamba-Instruct,68%,65%,100%,100%,90%,81%,-,-,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,68%,,
|
6 |
Cohere Command Text,59%,54%,100%,84%,78%,40%,-,-,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,59%,,
|
7 |
Claude 3 Haiku,86%,80%,100%,98%,95%,40%,-,-,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,86%,,
|
8 |
-
Gemini
|
9 |
SF-TextBase 70B,98%,63%,100%,90%,54%,8%,-,-,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,98%,,
|
10 |
SF-TextSum,82%,51%,100%,89%,87%,27%,-,-,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,82%,,
|
11 |
-
XGen
|
12 |
SF-TextBase 7B,82%,60%,100%,83%,69%,27%,-,-,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,82%,,
|
13 |
Mistral 7B,32%,42%,100%,97%,92%,82%,-,-,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,32%,0.426,
|
14 |
Mixtral 8x7B,89%,59%,100%,97%,71%,55%,-,-,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,89%,0.88,
|
15 |
LLaMA 3 8B,96%,76%,100%,99%,92%,85%,-,-,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,96%,0.598,
|
16 |
LLaMA 3 70B,98%,74%,100%,98%,83%,75%,-,-,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,98%,0.962,
|
17 |
-
Gemini
|
18 |
Claude 3 Opus,94%,81%,100%,96%,80%,56%,-,-,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,94%,,
|
19 |
Cohere Command R+,84%,56%,100%,97%,76%,45%,-,-,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,84%,,
|
|
|
5 |
AI21 Jamba-Instruct,68%,65%,100%,100%,90%,81%,-,-,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,68%,,
|
6 |
Cohere Command Text,59%,54%,100%,84%,78%,40%,-,-,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,59%,,
|
7 |
Claude 3 Haiku,86%,80%,100%,98%,95%,40%,-,-,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,86%,,
|
8 |
+
Gemini Pro 1,87%,74%,100%,92%,81%,48%,-,-,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,87%,,
|
9 |
SF-TextBase 70B,98%,63%,100%,90%,54%,8%,-,-,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,98%,,
|
10 |
SF-TextSum,82%,51%,100%,89%,87%,27%,-,-,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,82%,,
|
11 |
+
XGen 2,52%,52%,100%,56%,81%,51%,-,-,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,52%,,
|
12 |
SF-TextBase 7B,82%,60%,100%,83%,69%,27%,-,-,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,82%,,
|
13 |
Mistral 7B,32%,42%,100%,97%,92%,82%,-,-,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,32%,0.426,
|
14 |
Mixtral 8x7B,89%,59%,100%,97%,71%,55%,-,-,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,89%,0.88,
|
15 |
LLaMA 3 8B,96%,76%,100%,99%,92%,85%,-,-,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,96%,0.598,
|
16 |
LLaMA 3 70B,98%,74%,100%,98%,83%,75%,-,-,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,98%,0.962,
|
17 |
+
Gemini Pro 1.5,98%,81%,100%,97%,87%,69%,-,-,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,98%,,
|
18 |
Claude 3 Opus,94%,81%,100%,96%,80%,56%,-,-,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,94%,,
|
19 |
Cohere Command R+,84%,56%,100%,97%,76%,45%,-,-,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,84%,,
|
src/about.py
CHANGED
@@ -22,7 +22,7 @@ NUM_FEWSHOT = 0 # Change with your few shot
|
|
22 |
|
23 |
|
24 |
# Your leaderboard name
|
25 |
-
TITLE = """<h1 align="center" id="space-title"
|
26 |
<h3>Assess which LLMs are accurate enough or need fine-tuning, and weigh this versus tradeoffs of speed, costs, and trust and safety. This is based on human manual and automated evaluation with real operational CRM data per use case.</h3>
|
27 |
"""
|
28 |
|
@@ -83,7 +83,7 @@ CITATION_BUTTON_LABEL = "Copy the following snippet to cite these results"
|
|
83 |
CITATION_BUTTON_TEXT = r"""
|
84 |
@misc{crm-llm-leaderboard,
|
85 |
author = {Salesforce AI},
|
86 |
-
title = {
|
87 |
year = {2024},
|
88 |
publisher = {Salesforce AI},
|
89 |
howpublished = "\url{https://https://huggingface.co/spaces/Salesforce/crm_llm_leaderboard}"
|
|
|
22 |
|
23 |
|
24 |
# Your leaderboard name
|
25 |
+
TITLE = """<h1 align="center" id="space-title">Generative AI Leaderboard for CRM</h1>
|
26 |
<h3>Assess which LLMs are accurate enough or need fine-tuning, and weigh this versus tradeoffs of speed, costs, and trust and safety. This is based on human manual and automated evaluation with real operational CRM data per use case.</h3>
|
27 |
"""
|
28 |
|
|
|
83 |
CITATION_BUTTON_TEXT = r"""
|
84 |
@misc{crm-llm-leaderboard,
|
85 |
author = {Salesforce AI},
|
86 |
+
title = {Generative AI Leaderboard for CRM},
|
87 |
year = {2024},
|
88 |
publisher = {Salesforce AI},
|
89 |
howpublished = "\url{https://https://huggingface.co/spaces/Salesforce/crm_llm_leaderboard}"
|
src/display/utils.py
CHANGED
@@ -81,7 +81,8 @@ ts_eval_column_dict.append(["ts", ColumnContent, ColumnContent("Trust & Safety",
|
|
81 |
ts_eval_column_dict.append(["safety", ColumnContent, ColumnContent("Safety", "markdown", False)])
|
82 |
ts_eval_column_dict.append(["privacy", ColumnContent, ColumnContent("Privacy", "markdown", False)])
|
83 |
ts_eval_column_dict.append(["truthfulness", ColumnContent, ColumnContent("Truthfulness", "markdown", False)])
|
84 |
-
|
|
|
85 |
TSEvalColumn = make_dataclass("TSEvalColumn", ts_eval_column_dict, frozen=True)
|
86 |
|
87 |
|
|
|
81 |
ts_eval_column_dict.append(["safety", ColumnContent, ColumnContent("Safety", "markdown", False)])
|
82 |
ts_eval_column_dict.append(["privacy", ColumnContent, ColumnContent("Privacy", "markdown", False)])
|
83 |
ts_eval_column_dict.append(["truthfulness", ColumnContent, ColumnContent("Truthfulness", "markdown", False)])
|
84 |
+
ts_eval_column_dict.append(["crm_bias", ColumnContent, ColumnContent("CRM Bias", "markdown", False)])
|
85 |
+
# ts_eval_column_dict.append(["bias_no_ci", ColumnContent, ColumnContent("Bias No CI", "markdown", True)])
|
86 |
TSEvalColumn = make_dataclass("TSEvalColumn", ts_eval_column_dict, frozen=True)
|
87 |
|
88 |
|
src/populate.py
CHANGED
@@ -21,12 +21,15 @@ def get_leaderboard_df_crm(
|
|
21 |
leaderboard_cost_df = pd.read_csv(os.path.join(crm_results_path, "hf_leaderboard_latency_cost.csv"))
|
22 |
leaderboard_cost_df = leaderboard_cost_df[~leaderboard_cost_df["Model Name"].isin(sf_finetuned_models)]
|
23 |
leaderboard_cost_df = leaderboard_cost_df.join(ref_df.set_index("Model Name"), on="Model Name")
|
24 |
-
leaderboard_cost_df["LLM Provider"] = leaderboard_cost_df["LLM Provider"].fillna("Google")
|
25 |
leaderboard_cost_df = leaderboard_cost_df[cost_cols].round(decimals=2)
|
26 |
|
27 |
leaderboard_ts_df = pd.read_csv(os.path.join(crm_results_path, "hf_leaderboard_ts.csv"))
|
|
|
28 |
leaderboard_ts_df = leaderboard_ts_df[~leaderboard_ts_df["Model Name"].isin(sf_finetuned_models)]
|
29 |
leaderboard_ts_df = leaderboard_ts_df.join(ref_df.set_index("Model Name"), on="Model Name")
|
|
|
|
|
30 |
privacy_cols = leaderboard_ts_df[
|
31 |
[
|
32 |
"Privacy Zero-Shot Match Avoidance",
|
@@ -37,12 +40,14 @@ def get_leaderboard_df_crm(
|
|
37 |
].apply(lambda x: x.str.rstrip("%").astype("float") / 100.0, axis=1)
|
38 |
|
39 |
leaderboard_ts_df["Privacy"] = privacy_cols.mean(axis=1).transform(lambda x: "{:,.2%}".format(x))
|
|
|
40 |
|
41 |
ts_cols = leaderboard_ts_df[
|
42 |
[
|
43 |
"Safety",
|
44 |
"Privacy",
|
45 |
"Truthfulness",
|
|
|
46 |
]
|
47 |
].apply(lambda x: x.str.rstrip("%").astype("float") / 100.0, axis=1)
|
48 |
leaderboard_ts_df["Trust & Safety"] = ts_cols.mean(axis=1).transform(lambda x: "{:,.2%}".format(x))
|
|
|
21 |
leaderboard_cost_df = pd.read_csv(os.path.join(crm_results_path, "hf_leaderboard_latency_cost.csv"))
|
22 |
leaderboard_cost_df = leaderboard_cost_df[~leaderboard_cost_df["Model Name"].isin(sf_finetuned_models)]
|
23 |
leaderboard_cost_df = leaderboard_cost_df.join(ref_df.set_index("Model Name"), on="Model Name")
|
24 |
+
# leaderboard_cost_df["LLM Provider"] = leaderboard_cost_df["LLM Provider"].fillna("Google")
|
25 |
leaderboard_cost_df = leaderboard_cost_df[cost_cols].round(decimals=2)
|
26 |
|
27 |
leaderboard_ts_df = pd.read_csv(os.path.join(crm_results_path, "hf_leaderboard_ts.csv"))
|
28 |
+
leaderboard_ts__crm_bias_df = pd.read_csv(os.path.join(crm_results_path, "hf_leaderboard_crm_bias.csv"))
|
29 |
leaderboard_ts_df = leaderboard_ts_df[~leaderboard_ts_df["Model Name"].isin(sf_finetuned_models)]
|
30 |
leaderboard_ts_df = leaderboard_ts_df.join(ref_df.set_index("Model Name"), on="Model Name")
|
31 |
+
leaderboard_ts_df = leaderboard_ts_df.join(leaderboard_ts__crm_bias_df.set_index("Model Name"), on="Model Name")
|
32 |
+
# leaderboard_ts_df["LLM Provider"] = leaderboard_ts_df["LLM Provider"].fillna("Google")
|
33 |
privacy_cols = leaderboard_ts_df[
|
34 |
[
|
35 |
"Privacy Zero-Shot Match Avoidance",
|
|
|
40 |
].apply(lambda x: x.str.rstrip("%").astype("float") / 100.0, axis=1)
|
41 |
|
42 |
leaderboard_ts_df["Privacy"] = privacy_cols.mean(axis=1).transform(lambda x: "{:,.2%}".format(x))
|
43 |
+
leaderboard_ts_df["Bias No CI"] = leaderboard_ts_df["CRM Bias"].transform(lambda x: x.split(" ")[0])
|
44 |
|
45 |
ts_cols = leaderboard_ts_df[
|
46 |
[
|
47 |
"Safety",
|
48 |
"Privacy",
|
49 |
"Truthfulness",
|
50 |
+
"Bias No CI",
|
51 |
]
|
52 |
].apply(lambda x: x.str.rstrip("%").astype("float") / 100.0, axis=1)
|
53 |
leaderboard_ts_df["Trust & Safety"] = ts_cols.mean(axis=1).transform(lambda x: "{:,.2%}".format(x))
|