add filter: Use Case Type; add citation, add assumption text

app.py

@@ -2,6 +2,8 @@ import gradio as gr
 import pandas as pd
 
 from src.about import (  # CITATION_BUTTON_LABEL,; CITATION_BUTTON_TEXT,; EVALUATION_QUEUE_TEXT,
+    CITATION_BUTTON_LABEL,
+    CITATION_BUTTON_TEXT,
     INTRODUCTION_TEXT,
     LLM_BENCHMARKS_TEXT,
     TITLE,
@@ -29,7 +31,11 @@ leaderboard_df = original_df.copy()
 def update_table(
     hidden_df: pd.DataFrame,
     columns: list,
+    llm_query: list,
+    llm_provider_query: list,
     accuracy_method_query: str,
+    use_case_query: list,
+    use_case_type_query: list,
     # type_query: list,
     # precision_query: str,
     # size_query: list,
@@ -38,7 +44,14 @@ def update_table(
 ):
     # filtered_df = filter_models(hidden_df, type_query, size_query, precision_query, show_deleted)
     # filtered_df = filter_queries(query, filtered_df)
-    filtered_df = filter_accuracy_method_func(hidden_df, accuracy_method_query)
+    filtered_df = filter_llm_func(hidden_df, llm_query)
+    filtered_df = filter_llm_provider_func(filtered_df, llm_provider_query)
+    filtered_df = filter_accuracy_method_func(filtered_df, accuracy_method_query)
+
+    filtered_df["Use Case Area"] = filtered_df["Use Case Name"].apply(lambda x: x.split(": ")[0])
+    # print(filtered_df["Use Case Area"].unique())
+    filtered_df = filter_use_case_func(filtered_df, use_case_query)
+    filtered_df = filter_use_case_type_func(filtered_df, use_case_type_query)
     df = select_columns(filtered_df, columns)
     return df
 
@@ -47,6 +60,24 @@ def filter_accuracy_method_func(df: pd.DataFrame, accuracy_method_query: str) ->
     return df[df["Accuracy Method"] == accuracy_method_query]
 
 
+def filter_use_case_func(df: pd.DataFrame, use_case_query: list) -> pd.DataFrame:
+    # print(use_case_query)
+    # print(df[df["Use Case Name"].isin(["Service: Conversation summary"])])
+    return df[df["Use Case Name"].isin(use_case_query)]
+
+
+def filter_use_case_type_func(df: pd.DataFrame, use_case_type_query: list) -> pd.DataFrame:
+    return df[df["Use Case Type"].isin(use_case_type_query)]
+
+
+def filter_llm_func(df: pd.DataFrame, llm_query: list) -> pd.DataFrame:
+    return df[df["Model Name"].isin(llm_query)]
+
+
+def filter_llm_provider_func(df: pd.DataFrame, llm_provider_query: list) -> pd.DataFrame:
+    return df[df["LLM Provider"].isin(llm_provider_query)]
+
+
 # def search_table(df: pd.DataFrame, query: str) -> pd.DataFrame:
 #     return df[(df[AutoEvalColumn.model.name].str.contains(query, case=False))]
 
@@ -128,10 +159,6 @@ with demo:
                             elem_id="column-select",
                             interactive=True,
                         )
-                    # with gr.Row():
-                    #     deleted_models_visibility = gr.Checkbox(
-                    #         value=False, label="Show gated/private/deleted models", interactive=True
-                    #     )
                 # with gr.Column(min_width=320):
                 #     # with gr.Box(elem_id="box-filter"):
                 #     filter_columns_type = gr.CheckboxGroup(
@@ -141,21 +168,15 @@ with demo:
                 #         interactive=True,
                 #         elem_id="filter-columns-type",
                 #     )
-                # filter_columns_precision = gr.CheckboxGroup(
-                #     label="Precision",
-                #     choices=[i.value.name for i in Precision],
-                #     value=[i.value.name for i in Precision],
-                #     interactive=True,
-                #     elem_id="filter-columns-precision",
-                # )
-                # filter_columns_size = gr.CheckboxGroup(
-                #     label="Model sizes (in billions of parameters)",
-                #     choices=list(NUMERIC_INTERVALS.keys()),
-                #     value=list(NUMERIC_INTERVALS.keys()),
-                #     interactive=True,
-                #     elem_id="filter-columns-size",
-                # )
             with gr.Row():
+                with gr.Column():
+                    filter_use_case_type = gr.CheckboxGroup(
+                        choices=["Service", "Sales"],
+                        value=["Service", "Sales"],
+                        label="Use Case Area",
+                        info="",
+                        interactive=True,
+                    )
                 with gr.Column():
                     filter_use_case_type = gr.CheckboxGroup(
                         choices=["Summary", "Generation"],
@@ -167,7 +188,7 @@ with demo:
                 with gr.Column():
                     filter_use_case = gr.Dropdown(
                         choices=list(original_df["Use Case Name"].unique()),
-                        # value=list(original_df["Use Case Name"].unique()),
+                        value=list(original_df["Use Case Name"].unique()),
                         label="Use Case",
                         info="",
                         multiselect=True,
@@ -186,20 +207,20 @@ with demo:
                         choices=["Manual", "Auto"],
                         value="Manual",
                         label="Accuracy Method",
-                        info="accuracy method",
+                        info="",
                         interactive=True,
                     )
                 with gr.Column():
                     filter_accuracy_threshold = gr.Number(
                         value="3",
                         label="Accuracy Threshold",
-                        info="",
+                        info="Range: 0.0 to 4.0",
                         interactive=True,
                     )
                 with gr.Column():
                     filter_llm = gr.CheckboxGroup(
                         choices=list(original_df["Model Name"].unique()),
-                        value=list(original_df["Model Name"].unique()),
+                        value=list(leaderboard_df["Model Name"].unique()),
                         label="Model Name",
                         info="",
                         interactive=True,
@@ -207,7 +228,7 @@ with demo:
                 with gr.Column():
                     filter_llm_provider = gr.CheckboxGroup(
                         choices=list(original_df["LLM Provider"].unique()),
-                        value=list(original_df["LLM Provider"].unique()),
+                        value=list(leaderboard_df["LLM Provider"].unique()),
                         label="LLM Provider",
                         info="",
                         interactive=True,
@@ -244,7 +265,11 @@ with demo:
             # )
             for selector in [
                 shown_columns,
+                filter_llm,
+                filter_llm_provider,
                 filter_accuracy_method,
+                filter_use_case,
+                filter_use_case_type,
                 # filter_columns_type,
                 # filter_columns_precision,
                 # filter_columns_size,
@@ -255,7 +280,11 @@ with demo:
                     [
                         hidden_leaderboard_table_for_search,
                         shown_columns,
+                        filter_llm,
+                        filter_llm_provider,
                         filter_accuracy_method,
+                        filter_use_case,
+                        filter_use_case_type,
                         # filter_columns_type,
                         # filter_columns_precision,
                         # filter_columns_size,
@@ -269,15 +298,15 @@ with demo:
         with gr.TabItem("📝 About", elem_id="llm-benchmark-tab-table", id=2):
             gr.Markdown(LLM_BENCHMARKS_TEXT, elem_classes="markdown-text")
 
-    # with gr.Row():
-    #     with gr.Accordion("📙 Citation", open=False):
-    #         citation_button = gr.Textbox(
-    #             value=CITATION_BUTTON_TEXT,
-    #             label=CITATION_BUTTON_LABEL,
-    #             lines=20,
-    #             elem_id="citation-button",
-    #             show_copy_button=True,
-    #         )
+    with gr.Row():
+        with gr.Accordion("📙 Citation", open=False):
+            citation_button = gr.Textbox(
+                value=CITATION_BUTTON_TEXT,
+                label=CITATION_BUTTON_LABEL,
+                lines=20,
+                elem_id="citation-button",
+                show_copy_button=True,
+            )
 
 # scheduler = BackgroundScheduler()
 # scheduler.add_job(restart_space, "interval", seconds=1800)

crm-results/hf_leaderboard_accuracy.csv

@@ -1,202 +1,217 @@
-Use Case Name,Use Case Type,Accuracy Method,Model Name,Model Version,LLM Provider,Instruction Following,Completeness,Conciseness,Factuality,Average
-Service: Conversation summary,Summary,Auto,Gemini  Pro 1,Gemini  Pro 1,Google,3.9834791059280854,3.9193391642371234,3.847424684159378,3.871720116618076,3.9054907677356656
-Service: Conversation summary,Summary,Auto,GPT4-o,GPT4-o,OpenAI,3.9961127308066082,3.9844509232264333,3.9300291545189503,3.9669582118561713,3.9693877551020407
-Service: Conversation summary,Summary,Auto,AI21 Jamba-Instruct,AI21 (jamba-instruct-preview),AI21,3.998056365403304,3.9737609329446064,3.8104956268221573,3.9310009718172982,3.9283284742468414
-Service: Conversation summary,Summary,Auto,GPT 4 Turbo,gpt-4-0613,OpenAI,4.0,3.9805636540330416,3.9067055393586005,3.9504373177842567,3.9594266277939747
-Service: Conversation summary,Summary,Auto,SF-TextBase 70B,TextBase-70B (Llama FT),Salesforce,3.998056365403304,3.992225461613217,3.881438289601555,3.9591836734693877,3.957725947521866
-Service: Conversation summary,Summary,Auto,Claude 3 Haiku,Claude 3 Haiku,Anthropic,3.999028182701652,3.993197278911565,3.6997084548104957,3.9591836734693877,3.912779397473275
-Service: Conversation summary,Summary,Auto,Cohere Command R+,cohere.cmd-R+,Cohere AI,4.0,3.9951409135082603,3.8328474246841595,3.954324586977648,3.945578231292517
-Service: Conversation summary,Summary,Auto,XGen 22B,XGen 22B (1228),Salesforce,3.9941690962099123,3.836734693877551,3.9047619047619047,3.880466472303207,3.9040330417881437
-Service: Conversation summary,Summary,Auto,SF-TextBase 7B,TextBase-7B (Mistral FT),Salesforce,3.991253644314869,3.9407191448007777,3.7862001943634596,3.8746355685131197,3.898202137998057
-Service: Conversation summary,Summary,Auto,LLaMA 3 70B,Meta-Llama-3-70B-Instruct,Meta,3.999028182701652,3.990281827016521,3.9280855199222544,3.9591836734693877,3.9691448007774537
-Service: Conversation summary,Summary,Auto,Cohere Command Text,cohere.command-text-v14,Cohere AI,3.938775510204082,3.938775510204082,3.6530612244897958,3.7560738581146746,3.8216715257531586
-Service: Conversation summary,Summary,Auto,GPT 3.5 Turbo,gpt-3.5-turbo,OpenAI,3.9961127308066082,3.935860058309038,3.9834791059280854,3.9310009718172982,3.961613216715257
-Service: Conversation summary,Summary,Auto,Claude 3 Opus,Claude 3 (Opus),Anthropic,3.998056365403304,3.989310009718173,3.7657920310981536,3.9640427599611274,3.9293002915451893
-Service: Conversation summary,Summary,Auto,Gemini  Pro 1.5,Gemini  Pro 1.5,Google,3.9961127308066082,3.9727891156462585,3.943634596695821,3.9494655004859087,3.9655004859086493
-Service: Conversation summary,Summary,Auto,LLaMA 3 8B,Meta-Llama-3-8B-Instruct,Meta,4.0,3.9961127308066082,3.565597667638484,3.922254616132167,3.8709912536443145
-Service: Conversation summary,Summary,Auto,Mistral 7B,Mistral-7B-Instruct-v0.1,Mistral,3.927113702623907,3.924198250728863,3.327502429543246,3.7764820213799806,3.738824101068999
-Service: Conversation summary,Summary,Auto,Mixtral 8x7B,Mixtral-8x7B-v0.1,Mistral,3.987366375121477,3.9825072886297375,3.3751214771622933,3.9115646258503403,3.814139941690962
-Service: Conversation summary,Summary,Auto,SF-TextSum,Summarization model 7B for Service (Mistral FT),Salesforce,3.997084548104956,3.9280855199222544,3.7755102040816326,3.9037900874635567,3.9011175898931
-Sales: Email Generation,Generation,Auto,Mistral 7B,Mistral-7B-Instruct-v0.1,Mistral,1.9811320754716981,2.018867924528302,1.8867924528301887,2.2452830188679247,2.0330188679245285
-Sales: Email Generation,Generation,Auto,Gemini  Pro 1,Gemini  Pro 1,Google,3.207547169811321,3.0377358490566038,3.0754716981132075,3.188679245283019,3.1273584905660377
-Sales: Email Generation,Generation,Auto,LLaMA 3 8B,Meta-Llama-3-8B-Instruct,Meta,3.7358490566037736,3.0754716981132075,3.5849056603773586,3.660377358490566,3.5141509433962264
-Sales: Email Generation,Generation,Auto,SF-TextSum,Summarization model 7B for Service (Mistral FT),Salesforce,3.4150943396226414,2.9245283018867925,3.5283018867924527,3.4716981132075473,3.334905660377359
-Sales: Email Generation,Generation,Auto,Claude 3 Haiku,Claude 3 Haiku,Anthropic,2.5283018867924527,2.811320754716981,2.5283018867924527,3.0377358490566038,2.7264150943396226
-Sales: Email Generation,Generation,Auto,Cohere Command R+,cohere.cmd-R+,Cohere AI,3.6792452830188678,3.018867924528302,3.547169811320755,3.7358490566037736,3.4952830188679247
-Sales: Email Generation,Generation,Auto,Cohere Command Text,cohere.command-text-v14,Cohere AI,2.358490566037736,2.452830188679245,2.2641509433962264,2.7547169811320753,2.4575471698113205
-Sales: Email Generation,Generation,Auto,XGen 22B,XGen 22B (1228),Salesforce,2.339622641509434,2.2830188679245285,2.3962264150943398,2.509433962264151,2.3820754716981134
-Sales: Email Generation,Generation,Auto,SF-TextBase 70B,TextBase-70B (Llama FT),Salesforce,3.849056603773585,3.018867924528302,3.69811320754717,3.811320754716981,3.5943396226415096
-Sales: Email Generation,Generation,Auto,LLaMA 3 70B,Meta-Llama-3-70B-Instruct,Meta,3.849056603773585,3.0,3.7358490566037736,3.811320754716981,3.599056603773585
-Sales: Email Generation,Generation,Auto,Claude 3 Opus,Claude 3 (Opus),Anthropic,3.056603773584906,3.0,3.509433962264151,3.452830188679245,3.2547169811320753
-Sales: Email Generation,Generation,Auto,GPT4-o,GPT4-o,OpenAI,3.830188679245283,3.0377358490566038,3.4716981132075473,3.792452830188679,3.533018867924528
-Sales: Email Generation,Generation,Auto,Mixtral 8x7B,Mixtral-8x7B-v0.1,Mistral,2.9245283018867925,2.69811320754717,2.7547169811320753,3.0377358490566038,2.8537735849056607
-Sales: Email Generation,Generation,Auto,Gemini  Pro 1.5,Gemini  Pro 1.5,Google,3.7735849056603774,3.0,3.69811320754717,3.6792452830188678,3.537735849056604
-Sales: Email Generation,Generation,Auto,GPT 3.5 Turbo,gpt-3.5-turbo,OpenAI,3.056603773584906,2.69811320754717,3.056603773584906,3.4339622641509435,3.061320754716981
-Sales: Email Generation,Generation,Auto,GPT 4 Turbo,gpt-4-0613,OpenAI,3.547169811320755,3.0,3.30188679245283,3.7358490566037736,3.3962264150943398
-Sales: Email Generation,Generation,Auto,AI21 Jamba-Instruct,AI21 (jamba-instruct-preview),AI21,2.169811320754717,2.490566037735849,2.150943396226415,2.6226415094339623,2.3584905660377355
-Sales: Email Generation,Generation,Auto,SF-TextBase 7B,TextBase-7B (Mistral FT),Salesforce,3.150943396226415,2.8867924528301887,3.150943396226415,3.4716981132075473,3.165094339622642
-Service: Reply Recommendations,Generation,Auto,Mistral 7B,Mistral-7B-Instruct-v0.1,Mistral,3.3333333333333335,2.7222222222222223,2.9444444444444446,2.7777777777777777,2.9444444444444446
-Service: Reply Recommendations,Generation,Auto,Gemini  Pro 1,Gemini  Pro 1,Google,3.7222222222222223,3.5,2.611111111111111,3.1666666666666665,3.25
-Service: Reply Recommendations,Generation,Auto,GPT4-o,GPT4-o,OpenAI,3.9444444444444446,3.111111111111111,3.8333333333333335,3.388888888888889,3.5694444444444446
-Service: Reply Recommendations,Generation,Auto,SF-TextBase 70B,TextBase-70B (Llama FT),Salesforce,3.9444444444444446,3.3333333333333335,3.6666666666666665,3.2222222222222223,3.541666666666667
-Service: Reply Recommendations,Generation,Auto,LLaMA 3 8B,Meta-Llama-3-8B-Instruct,Meta,3.9444444444444446,3.111111111111111,3.5555555555555554,3.111111111111111,3.4305555555555554
-Service: Reply Recommendations,Generation,Auto,Claude 3 Haiku,Claude 3 Haiku,Anthropic,3.7777777777777777,3.1666666666666665,3.2777777777777777,3.111111111111111,3.333333333333333
-Service: Reply Recommendations,Generation,Auto,Cohere Command R+,cohere.cmd-R+,Cohere AI,3.611111111111111,3.1666666666666665,2.888888888888889,3.0,3.1666666666666665
-Service: Reply Recommendations,Generation,Auto,Cohere Command Text,cohere.command-text-v14,Cohere AI,3.2777777777777777,3.1666666666666665,2.5555555555555554,3.0555555555555554,3.013888888888889
-Service: Reply Recommendations,Generation,Auto,SF-TextBase 7B,TextBase-7B (Mistral FT),Salesforce,3.5555555555555554,2.8333333333333335,3.111111111111111,3.2777777777777777,3.1944444444444446
-Service: Reply Recommendations,Generation,Auto,XGen 22B,XGen 22B (1228),Salesforce,3.7222222222222223,3.111111111111111,3.2777777777777777,3.0,3.2777777777777777
-Service: Reply Recommendations,Generation,Auto,LLaMA 3 70B,Meta-Llama-3-70B-Instruct,Meta,4.0,3.2777777777777777,3.888888888888889,3.1666666666666665,3.583333333333333
-Service: Reply Recommendations,Generation,Auto,Claude 3 Opus,Claude 3 (Opus),Anthropic,4.0,3.5555555555555554,3.388888888888889,3.3333333333333335,3.5694444444444446
-Service: Reply Recommendations,Generation,Auto,Mixtral 8x7B,Mixtral-8x7B-v0.1,Mistral,3.888888888888889,3.1666666666666665,3.611111111111111,3.2222222222222223,3.4722222222222223
-Service: Reply Recommendations,Generation,Auto,Gemini  Pro 1.5,Gemini  Pro 1.5,Google,3.7777777777777777,2.888888888888889,3.8333333333333335,3.2777777777777777,3.4444444444444446
-Service: Reply Recommendations,Generation,Auto,GPT 3.5 Turbo,gpt-3.5-turbo,OpenAI,3.888888888888889,3.0555555555555554,3.7777777777777777,3.1666666666666665,3.472222222222222
-Service: Reply Recommendations,Generation,Auto,GPT 4 Turbo,gpt-4-0613,OpenAI,3.9444444444444446,3.1666666666666665,3.5,3.2222222222222223,3.458333333333333
-Service: Reply Recommendations,Generation,Auto,AI21 Jamba-Instruct,AI21 (jamba-instruct-preview),AI21,3.8333333333333335,3.2222222222222223,3.2777777777777777,3.0,3.333333333333333
-Service: Reply Recommendations,Generation,Auto,SF-TextSum,Summarization model 7B for Service (Mistral FT),Salesforce,3.888888888888889,3.1666666666666665,3.2777777777777777,3.2777777777777777,3.4027777777777777
-Sales & Service: Update CRM Info,Generation,Auto,Gemini  Pro 1,Gemini  Pro 1,Google,3.4,3.4242424242424243,3.4451219512195124,3.466666666666667,3.434007760532151
-Sales & Service: Update CRM Info,Generation,Auto,GPT4-o,GPT4-o,OpenAI,3.8848484848484848,3.6484848484848484,3.915151515151515,3.909090909090909,3.83939393939394
-Sales & Service: Update CRM Info,Generation,Auto,AI21 Jamba-Instruct,AI21 (jamba-instruct-preview),AI21,3.581818181818182,3.533333333333333,3.709090909090909,3.6787878787878787,3.625757575757576
-Sales & Service: Update CRM Info,Generation,Auto,GPT 4 Turbo,gpt-4-0613,OpenAI,3.709090909090909,3.6363636363636362,3.806060606060606,3.7333333333333334,3.7212121212121216
-Sales & Service: Update CRM Info,Generation,Auto,SF-TextBase 70B,TextBase-70B (Llama FT),Salesforce,3.824242424242424,3.775757575757576,3.83030303030303,3.806060606060606,3.809090909090909
-Sales & Service: Update CRM Info,Generation,Auto,Claude 3 Haiku,Claude 3 Haiku,Anthropic,3.618181818181818,3.624242424242424,3.7151515151515153,3.6666666666666665,3.656060606060606
-Sales & Service: Update CRM Info,Generation,Auto,Cohere Command R+,cohere.cmd-R+,Cohere AI,3.7636363636363637,3.624242424242424,3.757575757575758,3.812121212121212,3.7393939393939393
-Sales & Service: Update CRM Info,Generation,Auto,XGen 22B,XGen 22B (1228),Salesforce,3.7333333333333334,3.6787878787878787,3.7818181818181817,3.7454545454545456,3.734848484848485
-Sales & Service: Update CRM Info,Generation,Auto,SF-TextBase 7B,TextBase-7B (Mistral FT),Salesforce,2.9696969696969697,3.0545454545454547,2.8,3.096969696969697,2.9803030303030305
-Sales & Service: Update CRM Info,Generation,Auto,LLaMA 3 70B,Meta-Llama-3-70B-Instruct,Meta,3.806060606060606,3.7515151515151515,3.8181818181818183,3.812121212121212,3.796969696969697
-Sales & Service: Update CRM Info,Generation,Auto,Cohere Command Text,cohere.command-text-v14,Cohere AI,2.909090909090909,2.896969696969697,2.8666666666666667,2.903030303030303,2.893939393939394
-Sales & Service: Update CRM Info,Generation,Auto,GPT 3.5 Turbo,gpt-3.5-turbo,OpenAI,3.703030303030303,3.6606060606060606,3.7515151515151515,3.7515151515151515,3.716666666666667
-Sales & Service: Update CRM Info,Generation,Auto,Claude 3 Opus,Claude 3 (Opus),Anthropic,3.7454545454545456,3.606060606060606,3.812121212121212,3.7636363636363637,3.731818181818182
-Sales & Service: Update CRM Info,Generation,Auto,Gemini  Pro 1.5,Gemini  Pro 1.5,Google,3.478787878787879,3.3575757575757574,3.5393939393939395,3.503030303030303,3.4696969696969697
-Sales & Service: Update CRM Info,Generation,Auto,LLaMA 3 8B,Meta-Llama-3-8B-Instruct,Meta,2.787878787878788,3.242424242424242,2.3333333333333335,3.3575757575757574,2.93030303030303
-Sales & Service: Update CRM Info,Generation,Auto,Mistral 7B,Mistral-7B-Instruct-v0.1,Mistral,1.6606060606060606,1.6848484848484848,1.6424242424242423,1.6787878787878787,1.6666666666666667
-Sales & Service: Update CRM Info,Generation,Auto,Mixtral 8x7B,Mixtral-8x7B-v0.1,Mistral,3.6666666666666665,3.6363636363636362,3.690909090909091,3.690909090909091,3.6712121212121214
-Sales & Service: Update CRM Info,Generation,Auto,SF-TextSum,Summarization model 7B for Service (Mistral FT),Salesforce,2.981818181818182,3.139393939393939,3.012121212121212,3.1575757575757577,3.0727272727272728
-Service: Live Chat Summary,Summary,Auto,Gemini  Pro 1,Gemini  Pro 1,Google,3.542,3.479,3.303,3.381,3.42625
-Service: Live Chat Summary,Summary,Auto,GPT4-o,GPT4-o,OpenAI,3.965,3.436,3.946,3.641,3.747
-Service: Live Chat Summary,Summary,Auto,AI21 Jamba-Instruct,AI21 (jamba-instruct-preview),AI21,3.715,3.331,3.714,3.6236236236236237,3.595905905905906
-Service: Live Chat Summary,Summary,Auto,SF-TextBase 70B,TextBase-70B (Llama FT),Salesforce,3.882,3.4004004004004003,3.904,3.718,3.7261001001001
-Service: Live Chat Summary,Summary,Auto,Claude 3 Haiku,Claude 3 Haiku,Anthropic,3.927,3.566,3.607,3.588,3.6720000000000006
-Service: Live Chat Summary,Summary,Auto,Cohere Command R+,cohere.cmd-R+,Cohere AI,3.945,3.725,3.666,3.453,3.69725
-Service: Live Chat Summary,Summary,Auto,XGen 22B,XGen 22B (1228),Salesforce,3.703,3.247,3.689,3.605,3.561
-Service: Live Chat Summary,Summary,Auto,SF-TextBase 7B,TextBase-7B (Mistral FT),Salesforce,3.694,3.158,3.75,3.593,3.54875
-Service: Live Chat Summary,Summary,Auto,LLaMA 3 70B,Meta-Llama-3-70B-Instruct,Meta,3.903,3.309,3.947,3.765,3.731
-Service: Live Chat Summary,Summary,Auto,Cohere Command Text,cohere.command-text-v14,Cohere AI,3.744,3.497,3.533,3.502,3.569
-Service: Live Chat Summary,Summary,Auto,GPT 4 Turbo,gpt-4-0613,OpenAI,3.992,3.822,3.914,3.587,3.82875
-Service: Live Chat Summary,Summary,Auto,GPT 3.5 Turbo,gpt-3.5-turbo,OpenAI,3.835,3.164164164164164,3.891,3.717,3.651791041041041
-Service: Live Chat Summary,Summary,Auto,Claude 3 Opus,Claude 3 (Opus),Anthropic,3.862,3.492,3.864,3.574,3.698
-Service: Live Chat Summary,Summary,Auto,Gemini  Pro 1.5,Gemini  Pro 1.5,Google,3.724,3.635,3.567,3.396,3.5805
-Service: Live Chat Summary,Summary,Auto,LLaMA 3 8B,Meta-Llama-3-8B-Instruct,Meta,3.93,3.7,3.772,3.681,3.7707500000000005
-Service: Live Chat Summary,Summary,Auto,Mistral 7B,Mistral-7B-Instruct-v0.1,Mistral,3.942,3.353,3.831,3.492,3.6544999999999996
-Service: Live Chat Summary,Summary,Auto,Mixtral 8x7B,Mixtral-8x7B-v0.1,Mistral,3.866,3.379,3.846,3.609,3.6750000000000003
-Service: Live Chat Summary,Summary,Auto,SF-TextSum,Summarization model 7B for Service (Mistral FT),Salesforce,3.595,3.129129129129129,3.709,3.504,3.4842822822822823
-Service: Email Summary,Summary,Auto,Gemini  Pro 1,Gemini  Pro 1,Google,3.7755102040816326,3.7857142857142856,3.4285714285714284,3.561224489795918,3.6377551020408165
-Service: Email Summary,Summary,Auto,GPT4-o,GPT4-o,OpenAI,3.9381443298969074,3.9587628865979383,3.8762886597938144,3.752577319587629,3.8814432989690726
-Service: Email Summary,Summary,Auto,AI21 Jamba-Instruct,AI21 (jamba-instruct-preview),AI21,3.6122448979591835,3.5306122448979593,3.4285714285714284,3.377551020408163,3.4872448979591835
-Service: Email Summary,Summary,Auto,SF-TextBase 70B,TextBase-70B (Llama FT),Salesforce,3.989795918367347,3.979591836734694,3.9693877551020407,3.673469387755102,3.9030612244897958
-Service: Email Summary,Summary,Auto,Claude 3 Haiku,Claude 3 Haiku,Anthropic,3.979591836734694,3.9591836734693877,3.673469387755102,3.857142857142857,3.86734693877551
-Service: Email Summary,Summary,Auto,Cohere Command R+,cohere.cmd-R+,Cohere AI,4.0,4.0,3.857142857142857,3.693877551020408,3.8877551020408165
-Service: Email Summary,Summary,Auto,XGen 22B,XGen 22B (1228),Salesforce,3.5670103092783507,3.268041237113402,3.5463917525773194,3.288659793814433,3.4175257731958766
-Service: Email Summary,Summary,Auto,SF-TextBase 7B,TextBase-7B (Mistral FT),Salesforce,3.7755102040816326,3.8979591836734695,3.5816326530612246,3.479591836734694,3.683673469387755
-Service: Email Summary,Summary,Auto,LLaMA 3 70B,Meta-Llama-3-70B-Instruct,Meta,4.0,3.989795918367347,3.979591836734694,3.622448979591837,3.8979591836734695
-Service: Email Summary,Summary,Auto,Cohere Command Text,cohere.command-text-v14,Cohere AI,3.9591836734693877,3.836734693877551,3.7448979591836733,3.3877551020408165,3.732142857142857
-Service: Email Summary,Summary,Auto,GPT 4 Turbo,gpt-4-0613,OpenAI,3.989795918367347,4.0,3.8469387755102042,3.795918367346939,3.908163265306123
-Service: Email Summary,Summary,Auto,GPT 3.5 Turbo,gpt-3.5-turbo,OpenAI,3.989795918367347,3.86734693877551,3.9693877551020407,3.5918367346938775,3.854591836734694
-Service: Email Summary,Summary,Auto,Claude 3 Opus,Claude 3 (Opus),Anthropic,4.0,3.979591836734694,3.8469387755102042,3.795918367346939,3.9056122448979593
-Service: Email Summary,Summary,Auto,Gemini  Pro 1.5,Gemini  Pro 1.5,Google,3.989795918367347,3.9183673469387754,3.938775510204082,3.5510204081632653,3.849489795918367
-Service: Email Summary,Summary,Auto,LLaMA 3 8B,Meta-Llama-3-8B-Instruct,Meta,4.0,3.979591836734694,3.86734693877551,3.693877551020408,3.885204081632653
-Service: Email Summary,Summary,Auto,Mistral 7B,Mistral-7B-Instruct-v0.1,Mistral,3.8877551020408165,3.8979591836734695,3.520408163265306,3.642857142857143,3.737244897959184
-Service: Email Summary,Summary,Auto,Mixtral 8x7B,Mixtral-8x7B-v0.1,Mistral,3.979591836734694,3.9591836734693877,3.7346938775510203,3.673469387755102,3.836734693877551
-Service: Email Summary,Summary,Auto,SF-TextSum,Summarization model 7B for Service (Mistral FT),Salesforce,3.377551020408163,3.6530612244897958,3.36734693877551,3.377551020408163,3.443877551020408
-Service: Knowledge creation from Case Info,Generation,Auto,Gemini  Pro 1,Gemini  Pro 1,Google,3.1875,3.0625,2.8125,3.0,3.015625
-Service: Knowledge creation from Case Info,Generation,Auto,GPT4-o,GPT4-o,OpenAI,3.625,3.6875,3.1875,3.3125,3.453125
+Use Case Name,Use Case Type,Accuracy Method,Model Name,Model Version,LLM Provider,Factuality,Instruction Following,Conciseness,Completeness,Accuracy
+Service: Conversation summary,Summary,Auto,Gemini  Pro 1,Gemini  Pro 1,Google,3.871720116618076,3.9834791059280854,3.847424684159378,3.9193391642371234,3.9054907677356656
+Service: Conversation summary,Summary,Auto,GPT4-o,GPT4-o,OpenAI,3.9669582118561713,3.9961127308066082,3.9300291545189503,3.9844509232264333,3.9693877551020407
+Service: Conversation summary,Summary,Auto,AI21 Jamba-Instruct,AI21 (jamba-instruct-preview),AI21,3.9310009718172982,3.998056365403304,3.8104956268221573,3.9737609329446064,3.9283284742468414
+Service: Conversation summary,Summary,Auto,GPT 4 Turbo,gpt-4-0613,OpenAI,3.9504373177842567,4.0,3.9067055393586005,3.9805636540330416,3.9594266277939747
+Service: Conversation summary,Summary,Auto,SF-TextBase 70B,TextBase-70B (Llama FT),Salesforce,3.9591836734693877,3.998056365403304,3.881438289601555,3.992225461613217,3.957725947521866
+Service: Conversation summary,Summary,Auto,Claude 3 Haiku,Claude 3 Haiku,Anthropic,3.9591836734693877,3.999028182701652,3.6997084548104957,3.993197278911565,3.912779397473275
+Service: Conversation summary,Summary,Auto,Cohere Command R+,cohere.cmd-R+,Cohere AI,3.954324586977648,4.0,3.8328474246841595,3.9951409135082603,3.945578231292517
+Service: Conversation summary,Summary,Auto,XGen 22B,XGen 22B (1228),Salesforce,3.880466472303207,3.9941690962099123,3.9047619047619047,3.836734693877551,3.9040330417881437
+Service: Conversation summary,Summary,Auto,SF-TextBase 7B,TextBase-7B (Mistral FT),Salesforce,3.8746355685131197,3.991253644314869,3.7862001943634596,3.9407191448007777,3.898202137998057
+Service: Conversation summary,Summary,Auto,LLaMA 3 70B,Meta-Llama-3-70B-Instruct,Meta,3.9591836734693877,3.999028182701652,3.9280855199222544,3.990281827016521,3.9691448007774537
+Service: Conversation summary,Summary,Auto,Cohere Command Text,cohere.command-text-v14,Cohere AI,3.7560738581146746,3.938775510204082,3.6530612244897958,3.938775510204082,3.8216715257531586
+Service: Conversation summary,Summary,Auto,GPT 3.5 Turbo,gpt-3.5-turbo,OpenAI,3.9310009718172982,3.9961127308066082,3.9834791059280854,3.935860058309038,3.961613216715257
+Service: Conversation summary,Summary,Auto,Claude 3 Opus,Claude 3 (Opus),Anthropic,3.9640427599611274,3.998056365403304,3.7657920310981536,3.989310009718173,3.9293002915451893
+Service: Conversation summary,Summary,Auto,Gemini  Pro 1.5,Gemini  Pro 1.5,Google,3.9494655004859087,3.9961127308066082,3.943634596695821,3.9727891156462585,3.9655004859086493
+Service: Conversation summary,Summary,Auto,LLaMA 3 8B,Meta-Llama-3-8B-Instruct,Meta,3.922254616132167,4.0,3.565597667638484,3.9961127308066082,3.8709912536443145
+Service: Conversation summary,Summary,Auto,Mistral 7B,Mistral-7B-Instruct-v0.1,Mistral,3.7764820213799806,3.927113702623907,3.327502429543246,3.924198250728863,3.738824101068999
+Service: Conversation summary,Summary,Auto,Mixtral 8x7B,Mixtral-8x7B-v0.1,Mistral,3.9115646258503403,3.987366375121477,3.3751214771622933,3.9825072886297375,3.814139941690962
+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
+Sales: Email Generation,Generation,Auto,Mistral 7B,Mistral-7B-Instruct-v0.1,Mistral,2.2452830188679247,1.9811320754716981,1.8867924528301887,2.018867924528302,2.0330188679245285
+Sales: Email Generation,Generation,Auto,Gemini  Pro 1,Gemini  Pro 1,Google,3.188679245283019,3.207547169811321,3.0754716981132075,3.0377358490566038,3.1273584905660377
+Sales: Email Generation,Generation,Auto,LLaMA 3 8B,Meta-Llama-3-8B-Instruct,Meta,3.660377358490566,3.7358490566037736,3.5849056603773586,3.0754716981132075,3.5141509433962264
+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
+Sales: Email Generation,Generation,Auto,Claude 3 Haiku,Claude 3 Haiku,Anthropic,3.0377358490566038,2.5283018867924527,2.5283018867924527,2.811320754716981,2.7264150943396226
+Sales: Email Generation,Generation,Auto,Cohere Command R+,cohere.cmd-R+,Cohere AI,3.7358490566037736,3.6792452830188678,3.547169811320755,3.018867924528302,3.4952830188679247
+Sales: Email Generation,Generation,Auto,Cohere Command Text,cohere.command-text-v14,Cohere AI,2.7547169811320753,2.358490566037736,2.2641509433962264,2.452830188679245,2.4575471698113205
+Sales: Email Generation,Generation,Auto,XGen 22B,XGen 22B (1228),Salesforce,2.509433962264151,2.339622641509434,2.3962264150943398,2.2830188679245285,2.3820754716981134
+Sales: Email Generation,Generation,Auto,SF-TextBase 70B,TextBase-70B (Llama FT),Salesforce,3.811320754716981,3.849056603773585,3.69811320754717,3.018867924528302,3.5943396226415096
+Sales: Email Generation,Generation,Auto,LLaMA 3 70B,Meta-Llama-3-70B-Instruct,Meta,3.811320754716981,3.849056603773585,3.7358490566037736,3.0,3.599056603773585
+Sales: Email Generation,Generation,Auto,Claude 3 Opus,Claude 3 (Opus),Anthropic,3.452830188679245,3.056603773584906,3.509433962264151,3.0,3.2547169811320753
+Sales: Email Generation,Generation,Auto,GPT4-o,GPT4-o,OpenAI,3.792452830188679,3.830188679245283,3.4716981132075473,3.0377358490566038,3.533018867924528
+Sales: Email Generation,Generation,Auto,Mixtral 8x7B,Mixtral-8x7B-v0.1,Mistral,3.0377358490566038,2.9245283018867925,2.7547169811320753,2.69811320754717,2.8537735849056607
+Sales: Email Generation,Generation,Auto,Gemini  Pro 1.5,Gemini  Pro 1.5,Google,3.6792452830188678,3.7735849056603774,3.69811320754717,3.0,3.537735849056604
+Sales: Email Generation,Generation,Auto,GPT 3.5 Turbo,gpt-3.5-turbo,OpenAI,3.4339622641509435,3.056603773584906,3.056603773584906,2.69811320754717,3.061320754716981
+Sales: Email Generation,Generation,Auto,GPT 4 Turbo,gpt-4-0613,OpenAI,3.7358490566037736,3.547169811320755,3.30188679245283,3.0,3.3962264150943398
+Sales: Email Generation,Generation,Auto,AI21 Jamba-Instruct,AI21 (jamba-instruct-preview),AI21,2.6226415094339623,2.169811320754717,2.150943396226415,2.490566037735849,2.3584905660377355
+Sales: Email Generation,Generation,Auto,SF-TextBase 7B,TextBase-7B (Mistral FT),Salesforce,3.4716981132075473,3.150943396226415,3.150943396226415,2.8867924528301887,3.165094339622642
+Service: Reply Recommendations,Generation,Auto,Mistral 7B,Mistral-7B-Instruct-v0.1,Mistral,2.7777777777777777,3.3333333333333335,2.9444444444444446,2.7222222222222223,2.9444444444444446
+Service: Reply Recommendations,Generation,Auto,Gemini  Pro 1,Gemini  Pro 1,Google,3.1666666666666665,3.7222222222222223,2.611111111111111,3.5,3.25
+Service: Reply Recommendations,Generation,Auto,GPT4-o,GPT4-o,OpenAI,3.388888888888889,3.9444444444444446,3.8333333333333335,3.111111111111111,3.5694444444444446
+Service: Reply Recommendations,Generation,Auto,SF-TextBase 70B,TextBase-70B (Llama FT),Salesforce,3.2222222222222223,3.9444444444444446,3.6666666666666665,3.3333333333333335,3.541666666666667
+Service: Reply Recommendations,Generation,Auto,LLaMA 3 8B,Meta-Llama-3-8B-Instruct,Meta,3.111111111111111,3.9444444444444446,3.5555555555555554,3.111111111111111,3.4305555555555554
+Service: Reply Recommendations,Generation,Auto,Claude 3 Haiku,Claude 3 Haiku,Anthropic,3.111111111111111,3.7777777777777777,3.2777777777777777,3.1666666666666665,3.333333333333333
+Service: Reply Recommendations,Generation,Auto,Cohere Command R+,cohere.cmd-R+,Cohere AI,3.0,3.611111111111111,2.888888888888889,3.1666666666666665,3.1666666666666665
+Service: Reply Recommendations,Generation,Auto,Cohere Command Text,cohere.command-text-v14,Cohere AI,3.0555555555555554,3.2777777777777777,2.5555555555555554,3.1666666666666665,3.013888888888889
+Service: Reply Recommendations,Generation,Auto,SF-TextBase 7B,TextBase-7B (Mistral FT),Salesforce,3.2777777777777777,3.5555555555555554,3.111111111111111,2.8333333333333335,3.1944444444444446
+Service: Reply Recommendations,Generation,Auto,XGen 22B,XGen 22B (1228),Salesforce,3.0,3.7222222222222223,3.2777777777777777,3.111111111111111,3.2777777777777777
+Service: Reply Recommendations,Generation,Auto,LLaMA 3 70B,Meta-Llama-3-70B-Instruct,Meta,3.1666666666666665,4.0,3.888888888888889,3.2777777777777777,3.583333333333333
+Service: Reply Recommendations,Generation,Auto,Claude 3 Opus,Claude 3 (Opus),Anthropic,3.3333333333333335,4.0,3.388888888888889,3.5555555555555554,3.5694444444444446
+Service: Reply Recommendations,Generation,Auto,Mixtral 8x7B,Mixtral-8x7B-v0.1,Mistral,3.2222222222222223,3.888888888888889,3.611111111111111,3.1666666666666665,3.4722222222222223
+Service: Reply Recommendations,Generation,Auto,Gemini  Pro 1.5,Gemini  Pro 1.5,Google,3.2777777777777777,3.7777777777777777,3.8333333333333335,2.888888888888889,3.4444444444444446
+Service: Reply Recommendations,Generation,Auto,GPT 3.5 Turbo,gpt-3.5-turbo,OpenAI,3.1666666666666665,3.888888888888889,3.7777777777777777,3.0555555555555554,3.472222222222222
+Service: Reply Recommendations,Generation,Auto,GPT 4 Turbo,gpt-4-0613,OpenAI,3.2222222222222223,3.9444444444444446,3.5,3.1666666666666665,3.458333333333333
+Service: Reply Recommendations,Generation,Auto,AI21 Jamba-Instruct,AI21 (jamba-instruct-preview),AI21,3.0,3.8333333333333335,3.2777777777777777,3.2222222222222223,3.333333333333333
+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
+Sales & Service: Update CRM Info,Generation,Auto,Gemini  Pro 1,Gemini  Pro 1,Google,3.466666666666667,3.4,3.4451219512195124,3.4242424242424243,3.434007760532151
+Sales & Service: Update CRM Info,Generation,Auto,GPT4-o,GPT4-o,OpenAI,3.909090909090909,3.8848484848484848,3.915151515151515,3.6484848484848484,3.83939393939394
+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
+Sales & Service: Update CRM Info,Generation,Auto,GPT 4 Turbo,gpt-4-0613,OpenAI,3.7333333333333334,3.709090909090909,3.806060606060606,3.6363636363636362,3.7212121212121216
+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
+Sales & Service: Update CRM Info,Generation,Auto,Claude 3 Haiku,Claude 3 Haiku,Anthropic,3.6666666666666665,3.618181818181818,3.7151515151515153,3.624242424242424,3.656060606060606
+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
+Sales & Service: Update CRM Info,Generation,Auto,XGen 22B,XGen 22B (1228),Salesforce,3.7454545454545456,3.7333333333333334,3.7818181818181817,3.6787878787878787,3.734848484848485
+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
+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
+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
+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
+Sales & Service: Update CRM Info,Generation,Auto,Claude 3 Opus,Claude 3 (Opus),Anthropic,3.7636363636363637,3.7454545454545456,3.812121212121212,3.606060606060606,3.731818181818182
+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
+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
+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
+Sales & Service: Update CRM Info,Generation,Auto,Mixtral 8x7B,Mixtral-8x7B-v0.1,Mistral,3.690909090909091,3.6666666666666665,3.690909090909091,3.6363636363636362,3.6712121212121214
+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
+Service: Live Chat Summary,Summary,Auto,Gemini  Pro 1,Gemini  Pro 1,Google,3.381,3.542,3.303,3.479,3.42625
+Service: Live Chat Summary,Summary,Auto,GPT4-o,GPT4-o,OpenAI,3.641,3.965,3.946,3.436,3.747
+Service: Live Chat Summary,Summary,Auto,AI21 Jamba-Instruct,AI21 (jamba-instruct-preview),AI21,3.6236236236236237,3.715,3.714,3.331,3.595905905905906
+Service: Live Chat Summary,Summary,Auto,SF-TextBase 70B,TextBase-70B (Llama FT),Salesforce,3.718,3.882,3.904,3.4004004004004003,3.7261001001001
+Service: Live Chat Summary,Summary,Auto,Claude 3 Haiku,Claude 3 Haiku,Anthropic,3.588,3.927,3.607,3.566,3.6720000000000006
+Service: Live Chat Summary,Summary,Auto,Cohere Command R+,cohere.cmd-R+,Cohere AI,3.453,3.945,3.666,3.725,3.69725
+Service: Live Chat Summary,Summary,Auto,XGen 22B,XGen 22B (1228),Salesforce,3.605,3.703,3.689,3.247,3.561
+Service: Live Chat Summary,Summary,Auto,SF-TextBase 7B,TextBase-7B (Mistral FT),Salesforce,3.593,3.694,3.75,3.158,3.54875
+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
+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
+Service: Live Chat Summary,Summary,Auto,GPT 4 Turbo,gpt-4-0613,OpenAI,3.587,3.992,3.914,3.822,3.82875
+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
+Service: Live Chat Summary,Summary,Auto,Claude 3 Opus,Claude 3 (Opus),Anthropic,3.574,3.862,3.864,3.492,3.698
+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
+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
+Service: Live Chat Summary,Summary,Auto,Mistral 7B,Mistral-7B-Instruct-v0.1,Mistral,3.492,3.942,3.831,3.353,3.6544999999999996
+Service: Live Chat Summary,Summary,Auto,Mixtral 8x7B,Mixtral-8x7B-v0.1,Mistral,3.609,3.866,3.846,3.379,3.6750000000000003
+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
+Service: Email Summary,Summary,Auto,Gemini  Pro 1,Gemini  Pro 1,Google,3.561224489795918,3.7755102040816326,3.4285714285714284,3.7857142857142856,3.6377551020408165
+Service: Email Summary,Summary,Auto,GPT4-o,GPT4-o,OpenAI,3.752577319587629,3.9381443298969074,3.8762886597938144,3.9587628865979383,3.8814432989690726
+Service: Email Summary,Summary,Auto,AI21 Jamba-Instruct,AI21 (jamba-instruct-preview),AI21,3.377551020408163,3.6122448979591835,3.4285714285714284,3.5306122448979593,3.4872448979591835
+Service: Email Summary,Summary,Auto,SF-TextBase 70B,TextBase-70B (Llama FT),Salesforce,3.673469387755102,3.989795918367347,3.9693877551020407,3.979591836734694,3.9030612244897958
+Service: Email Summary,Summary,Auto,Claude 3 Haiku,Claude 3 Haiku,Anthropic,3.857142857142857,3.979591836734694,3.673469387755102,3.9591836734693877,3.86734693877551
+Service: Email Summary,Summary,Auto,Cohere Command R+,cohere.cmd-R+,Cohere AI,3.693877551020408,4.0,3.857142857142857,4.0,3.8877551020408165
+Service: Email Summary,Summary,Auto,XGen 22B,XGen 22B (1228),Salesforce,3.288659793814433,3.5670103092783507,3.5463917525773194,3.268041237113402,3.4175257731958766
+Service: Email Summary,Summary,Auto,SF-TextBase 7B,TextBase-7B (Mistral FT),Salesforce,3.479591836734694,3.7755102040816326,3.5816326530612246,3.8979591836734695,3.683673469387755
+Service: Email Summary,Summary,Auto,LLaMA 3 70B,Meta-Llama-3-70B-Instruct,Meta,3.622448979591837,4.0,3.979591836734694,3.989795918367347,3.8979591836734695
+Service: Email Summary,Summary,Auto,Cohere Command Text,cohere.command-text-v14,Cohere AI,3.3877551020408165,3.9591836734693877,3.7448979591836733,3.836734693877551,3.732142857142857
+Service: Email Summary,Summary,Auto,GPT 4 Turbo,gpt-4-0613,OpenAI,3.795918367346939,3.989795918367347,3.8469387755102042,4.0,3.908163265306123
+Service: Email Summary,Summary,Auto,GPT 3.5 Turbo,gpt-3.5-turbo,OpenAI,3.5918367346938775,3.989795918367347,3.9693877551020407,3.86734693877551,3.854591836734694
+Service: Email Summary,Summary,Auto,Claude 3 Opus,Claude 3 (Opus),Anthropic,3.795918367346939,4.0,3.8469387755102042,3.979591836734694,3.9056122448979593
+Service: Email Summary,Summary,Auto,Gemini  Pro 1.5,Gemini  Pro 1.5,Google,3.5510204081632653,3.989795918367347,3.938775510204082,3.9183673469387754,3.849489795918367
+Service: Email Summary,Summary,Auto,LLaMA 3 8B,Meta-Llama-3-8B-Instruct,Meta,3.693877551020408,4.0,3.86734693877551,3.979591836734694,3.885204081632653
+Service: Email Summary,Summary,Auto,Mistral 7B,Mistral-7B-Instruct-v0.1,Mistral,3.642857142857143,3.8877551020408165,3.520408163265306,3.8979591836734695,3.737244897959184
+Service: Email Summary,Summary,Auto,Mixtral 8x7B,Mixtral-8x7B-v0.1,Mistral,3.673469387755102,3.979591836734694,3.7346938775510203,3.9591836734693877,3.836734693877551
+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
+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
+Service: Knowledge creation from Case Info,Generation,Auto,GPT4-o,GPT4-o,OpenAI,3.3125,3.625,3.1875,3.6875,3.453125
 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
-Service: Knowledge creation from Case Info,Generation,Auto,SF-TextBase 70B,TextBase-70B (Llama FT),Salesforce,3.5625,3.4375,3.25,3.1875,3.359375
-Service: Knowledge creation from Case Info,Generation,Auto,Claude 3 Haiku,Claude 3 Haiku,Anthropic,3.5625,3.25,3.1875,3.125,3.28125
-Service: Knowledge creation from Case Info,Generation,Auto,Cohere Command R+,cohere.cmd-R+,Cohere AI,3.5625,3.1875,3.125,3.0,3.21875
-Service: Knowledge creation from Case Info,Generation,Auto,XGen 22B,XGen 22B (1228),Salesforce,3.25,3.1875,3.0,3.1875,3.15625
-Service: Knowledge creation from Case Info,Generation,Auto,SF-TextBase 7B,TextBase-7B (Mistral FT),Salesforce,3.5625,3.375,3.0625,3.3125,3.328125
-Service: Knowledge creation from Case Info,Generation,Auto,LLaMA 3 70B,Meta-Llama-3-70B-Instruct,Meta,2.4375,2.3125,2.0625,3.0625,2.46875
-Service: Knowledge creation from Case Info,Generation,Auto,Cohere Command Text,cohere.command-text-v14,Cohere AI,3.0625,3.0625,2.8125,3.125,3.015625
-Service: Knowledge creation from Case Info,Generation,Auto,GPT 4 Turbo,gpt-4-0613,OpenAI,3.6875,3.375,3.1875,3.3125,3.390625
-Service: Knowledge creation from Case Info,Generation,Auto,GPT 3.5 Turbo,gpt-3.5-turbo,OpenAI,3.5625,3.4375,3.3125,3.25,3.390625
-Service: Knowledge creation from Case Info,Generation,Auto,Claude 3 Opus,Claude 3 (Opus),Anthropic,3.375,3.1875,2.9375,3.125,3.15625
-Service: Knowledge creation from Case Info,Generation,Auto,Gemini  Pro 1.5,Gemini  Pro 1.5,Google,3.4375,3.125,3.125,3.1875,3.21875
-Service: Knowledge creation from Case Info,Generation,Auto,LLaMA 3 8B,Meta-Llama-3-8B-Instruct,Meta,3.5625,3.4375,3.0,3.1875,3.296875
-Service: Knowledge creation from Case Info,Generation,Auto,Mistral 7B,Mistral-7B-Instruct-v0.1,Mistral,3.5625,3.375,2.9375,3.0625,3.234375
-Service: Knowledge creation from Case Info,Generation,Auto,Mixtral 8x7B,Mixtral-8x7B-v0.1,Mistral,3.4375,3.375,3.125,3.25,3.296875
-Service: Knowledge creation from Case Info,Generation,Auto,SF-TextSum,Summarization model 7B for Service (Mistral FT),Salesforce,3.4375,3.3125,3.0,3.125,3.21875
-Sales: Email Summary,Summary,Auto,Gemini  Pro 1,Gemini  Pro 1,Google,3.3877551020408165,3.4285714285714284,3.193877551020408,3.377551020408163,3.346938775510204
-Sales: Email Summary,Summary,Auto,GPT4-o,GPT4-o,OpenAI,3.9693877551020407,3.8979591836734695,3.86734693877551,3.8877551020408165,3.9056122448979593
-Sales: Email Summary,Summary,Auto,AI21 Jamba-Instruct,AI21 (jamba-instruct-preview),AI21,3.8979591836734695,3.663265306122449,3.9183673469387754,3.5714285714285716,3.7627551020408165
-Sales: Email Summary,Summary,Auto,SF-TextBase 70B,TextBase-70B (Llama FT),Salesforce,3.979591836734694,3.836734693877551,3.9591836734693877,3.663265306122449,3.8596938775510203
-Sales: Email Summary,Summary,Auto,Claude 3 Haiku,Claude 3 Haiku,Anthropic,3.989795918367347,3.7653061224489797,3.9183673469387754,3.857142857142857,3.88265306122449
-Sales: Email Summary,Summary,Auto,Cohere Command R+,cohere.cmd-R+,Cohere AI,3.8979591836734695,3.7551020408163267,3.795918367346939,3.7244897959183674,3.793367346938776
-Sales: Email Summary,Summary,Auto,XGen 22B,XGen 22B (1228),Salesforce,2.979381443298969,2.8969072164948453,3.142857142857143,3.2783505154639174,3.0743740795287184
-Sales: Email Summary,Summary,Auto,SF-TextBase 7B,TextBase-7B (Mistral FT),Salesforce,3.6020408163265305,3.510204081632653,3.663265306122449,3.357142857142857,3.5331632653061225
-Sales: Email Summary,Summary,Auto,LLaMA 3 70B,Meta-Llama-3-70B-Instruct,Meta,3.9693877551020407,3.8979591836734695,3.9591836734693877,3.642857142857143,3.86734693877551
-Sales: Email Summary,Summary,Auto,Cohere Command Text,cohere.command-text-v14,Cohere AI,2.9690721649484537,3.2783505154639174,2.649484536082474,3.154639175257732,3.0128865979381443
-Sales: Email Summary,Summary,Auto,GPT 4 Turbo,gpt-4-0613,OpenAI,4.0,3.9591836734693877,3.9285714285714284,3.826530612244898,3.928571428571429
-Sales: Email Summary,Summary,Auto,GPT 3.5 Turbo,gpt-3.5-turbo,OpenAI,3.9591836734693877,3.693877551020408,3.9693877551020407,3.7857142857142856,3.8520408163265305
-Sales: Email Summary,Summary,Auto,Claude 3 Opus,Claude 3 (Opus),Anthropic,4.0,3.938775510204082,3.9183673469387754,3.826530612244898,3.920918367346939
-Sales: Email Summary,Summary,Auto,Gemini  Pro 1.5,Gemini  Pro 1.5,Google,3.9285714285714284,3.663265306122449,3.9183673469387754,3.7653061224489797,3.8188775510204085
-Sales: Email Summary,Summary,Auto,LLaMA 3 8B,Meta-Llama-3-8B-Instruct,Meta,4.0,3.806122448979592,3.9693877551020407,3.7346938775510203,3.877551020408163
-Sales: Email Summary,Summary,Auto,Mistral 7B,Mistral-7B-Instruct-v0.1,Mistral,3.948453608247423,3.948453608247423,3.7653061224489797,3.6288659793814433,3.822769829581317
-Sales: Email Summary,Summary,Auto,Mixtral 8x7B,Mixtral-8x7B-v0.1,Mistral,3.9591836734693877,3.816326530612245,3.8469387755102042,3.5306122448979593,3.788265306122449
-Sales: Email Summary,Summary,Auto,SF-TextSum,Summarization model 7B for Service (Mistral FT),Salesforce,3.836734693877551,3.642857142857143,3.7346938775510203,3.5306122448979593,3.686224489795918
-Service: Call Summary,Summary,Auto,Mistral 7B,Mistral-7B-Instruct-v0.1,Mistral,3.95,3.925,3.925,3.75,3.8875
-Service: Call Summary,Summary,Auto,Gemini  Pro 1,Gemini  Pro 1,Google,3.95,3.85,3.9,3.775,3.8687500000000004
-Service: Call Summary,Summary,Auto,GPT4-o,GPT4-o,OpenAI,3.95,3.925,3.95,3.925,3.9375
-Service: Call Summary,Summary,Auto,SF-TextBase 70B,TextBase-70B (Llama FT),Salesforce,3.95,3.95,3.95,3.975,3.9562500000000003
-Service: Call Summary,Summary,Auto,LLaMA 3 8B,Meta-Llama-3-8B-Instruct,Meta,3.95,3.95,3.95,3.9,3.9375000000000004
-Service: Call Summary,Summary,Auto,Claude 3 Haiku,Claude 3 Haiku,Anthropic,3.925,3.925,3.775,3.8,3.85625
-Service: Call Summary,Summary,Auto,Cohere Command R+,cohere.cmd-R+,Cohere AI,3.95,3.925,3.95,3.925,3.9375
-Service: Call Summary,Summary,Auto,Cohere Command Text,cohere.command-text-v14,Cohere AI,3.75,3.85,3.675,3.6,3.7187499999999996
-Service: Call Summary,Summary,Auto,SF-TextBase 7B,TextBase-7B (Mistral FT),Salesforce,4.0,3.875,3.975,3.95,3.95
-Service: Call Summary,Summary,Auto,XGen 22B,XGen 22B (1228),Salesforce,3.95,3.8,3.95,3.65,3.8375
-Service: Call Summary,Summary,Auto,LLaMA 3 70B,Meta-Llama-3-70B-Instruct,Meta,3.95,3.925,3.95,3.875,3.925
-Service: Call Summary,Summary,Auto,Claude 3 Opus,Claude 3 (Opus),Anthropic,4.0,4.0,4.0,3.95,3.9875
-Service: Call Summary,Summary,Auto,Mixtral 8x7B,Mixtral-8x7B-v0.1,Mistral,3.975,3.925,3.95,3.825,3.91875
-Service: Call Summary,Summary,Auto,Gemini  Pro 1.5,Gemini  Pro 1.5,Google,3.95,3.925,4.0,3.95,3.95625
-Service: Call Summary,Summary,Auto,GPT 3.5 Turbo,gpt-3.5-turbo,OpenAI,3.95,3.875,3.95,3.85,3.90625
-Service: Call Summary,Summary,Auto,GPT 4 Turbo,gpt-4-0613,OpenAI,3.95,3.925,3.95,3.9,3.93125
-Service: Call Summary,Summary,Auto,AI21 Jamba-Instruct,AI21 (jamba-instruct-preview),AI21,3.95,3.9,3.925,3.825,3.8999999999999995
-Service: Call Summary,Summary,Auto,SF-TextSum,Summarization model 7B for Service (Mistral FT),Salesforce,3.95,3.875,3.95,3.875,3.9125
-Sales: Call Summary,Summary,Auto,Gemini  Pro 1,Gemini  Pro 1,Google,3.0526315789473686,2.8947368421052633,2.473684210526316,2.789473684210526,2.8026315789473686
-Sales: Call Summary,Summary,Auto,GPT4-o,GPT4-o,OpenAI,3.9473684210526314,3.210526315789474,3.473684210526316,3.9473684210526314,3.6447368421052633
-Sales: Call Summary,Summary,Auto,AI21 Jamba-Instruct,AI21 (jamba-instruct-preview),AI21,3.3157894736842106,3.0,3.1052631578947367,3.3684210526315788,3.1973684210526314
-Sales: Call Summary,Summary,Auto,SF-TextBase 70B,TextBase-70B (Llama FT),Salesforce,3.3684210526315788,2.789473684210526,3.263157894736842,3.6842105263157894,3.276315789473684
-Sales: Call Summary,Summary,Auto,Claude 3 Haiku,Claude 3 Haiku,Anthropic,3.473684210526316,3.0526315789473686,3.3157894736842106,3.6315789473684212,3.368421052631579
-Sales: Call Summary,Summary,Auto,Cohere Command R+,cohere.cmd-R+,Cohere AI,3.6842105263157894,3.0526315789473686,2.8421052631578947,3.789473684210526,3.3421052631578942
-Sales: Call Summary,Summary,Auto,XGen 22B,XGen 22B (1228),Salesforce,2.736842105263158,2.6315789473684212,2.5789473684210527,2.789473684210526,2.6842105263157894
-Sales: Call Summary,Summary,Auto,SF-TextBase 7B,TextBase-7B (Mistral FT),Salesforce,3.1052631578947367,2.6315789473684212,3.210526315789474,3.210526315789474,3.0394736842105265
-Sales: Call Summary,Summary,Auto,LLaMA 3 70B,Meta-Llama-3-70B-Instruct,Meta,3.263157894736842,2.8421052631578947,3.473684210526316,3.4210526315789473,3.2499999999999996
-Sales: Call Summary,Summary,Auto,Cohere Command Text,cohere.command-text-v14,Cohere AI,2.473684210526316,2.3157894736842106,2.1052631578947367,1.7894736842105263,2.1710526315789473
-Sales: Call Summary,Summary,Auto,GPT 4 Turbo,gpt-4-0613,OpenAI,3.736842105263158,3.0,3.3157894736842106,4.0,3.513157894736842
-Sales: Call Summary,Summary,Auto,GPT 3.5 Turbo,gpt-3.5-turbo,OpenAI,3.3684210526315788,2.888888888888889,3.8421052631578947,3.8421052631578947,3.485380116959064
-Sales: Call Summary,Summary,Auto,Claude 3 Opus,Claude 3 (Opus),Anthropic,3.8947368421052633,3.0,3.5789473684210527,3.9473684210526314,3.605263157894737
-Sales: Call Summary,Summary,Auto,Gemini  Pro 1.5,Gemini  Pro 1.5,Google,3.6315789473684212,3.0526315789473686,3.0526315789473686,3.736842105263158,3.3684210526315788
-Sales: Call Summary,Summary,Auto,LLaMA 3 8B,Meta-Llama-3-8B-Instruct,Meta,3.1578947368421053,2.8947368421052633,2.6315789473684212,2.9473684210526314,2.9078947368421053
-Sales: Call Summary,Summary,Auto,Mistral 7B,Mistral-7B-Instruct-v0.1,Mistral,3.4210526315789473,3.1052631578947367,3.210526315789474,3.5555555555555554,3.323099415204678
-Sales: Call Summary,Summary,Auto,Mixtral 8x7B,Mixtral-8x7B-v0.1,Mistral,3.263157894736842,3.0,2.6315789473684212,3.263157894736842,3.0394736842105265
-Sales: Call Summary,Summary,Auto,SF-TextSum,Summarization model 7B for Service (Mistral FT),Salesforce,2.9473684210526314,2.736842105263158,3.0,3.0526315789473686,2.9342105263157894
-Service: Live Chat Insights,Summary,Auto,Gemini  Pro 1,Gemini  Pro 1,Google,3.97265625,3.8359375,3.87109375,3.76953125,3.8623046875
-Service: Live Chat Insights,Summary,Auto,GPT4-o,GPT4-o,OpenAI,4.0,3.91796875,3.9765625,3.91015625,3.951171875
-Service: Live Chat Insights,Summary,Auto,AI21 Jamba-Instruct,AI21 (jamba-instruct-preview),AI21,3.97265625,3.8203125,3.8515625,3.859375,3.8759765625
-Service: Live Chat Insights,Summary,Auto,SF-TextBase 70B,TextBase-70B (Llama FT),Salesforce,3.99609375,3.96875,3.94140625,3.85546875,3.9404296875
-Service: Live Chat Insights,Summary,Auto,Claude 3 Haiku,Claude 3 Haiku,Anthropic,4.0,3.98046875,3.796875,3.8828125,3.9150390625
-Service: Live Chat Insights,Summary,Auto,Cohere Command R+,cohere.cmd-R+,Cohere AI,3.984375,3.9375,3.8671875,3.87109375,3.9150390625
-Service: Live Chat Insights,Summary,Auto,XGen 22B,XGen 22B (1228),Salesforce,3.9765625,3.4375,3.921875,3.74609375,3.7705078125
-Service: Live Chat Insights,Summary,Auto,SF-TextBase 7B,TextBase-7B (Mistral FT),Salesforce,3.97265625,3.6875,3.90625,3.71484375,3.8203125
-Service: Live Chat Insights,Summary,Auto,LLaMA 3 70B,Meta-Llama-3-70B-Instruct,Meta,3.9921875,3.93359375,3.97265625,3.85546875,3.9384765625
-Service: Live Chat Insights,Summary,Auto,Cohere Command Text,cohere.command-text-v14,Cohere AI,3.9375,3.8671875,3.484375,3.51171875,3.7001953125
-Service: Live Chat Insights,Summary,Auto,GPT 4 Turbo,gpt-4-0613,OpenAI,3.9921875,3.9375,3.93359375,3.91015625,3.943359375
-Service: Live Chat Insights,Summary,Auto,GPT 3.5 Turbo,gpt-3.5-turbo,OpenAI,3.96875,3.71875,4.0,3.90234375,3.8974609375
-Service: Live Chat Insights,Summary,Auto,Claude 3 Opus,Claude 3 (Opus),Anthropic,4.0,3.98046875,3.82421875,3.9140625,3.9296875
-Service: Live Chat Insights,Summary,Auto,Gemini  Pro 1.5,Gemini  Pro 1.5,Google,3.98828125,3.88671875,3.9375,3.77734375,3.8974609375
-Service: Live Chat Insights,Summary,Auto,LLaMA 3 8B,Meta-Llama-3-8B-Instruct,Meta,3.9921875,3.96484375,3.62890625,3.8359375,3.85546875
-Service: Live Chat Insights,Summary,Auto,Mistral 7B,Mistral-7B-Instruct-v0.1,Mistral,3.9921875,3.875,3.7734375,3.8125,3.86328125
-Service: Live Chat Insights,Summary,Auto,Mixtral 8x7B,Mixtral-8x7B-v0.1,Mistral,3.96875,3.953125,3.6015625,3.8046875,3.83203125
-Service: Live Chat Insights,Summary,Auto,SF-TextSum,Summarization model 7B for Service (Mistral FT),Salesforce,3.9765625,3.7734375,3.92578125,3.80859375,3.87109375
+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
+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
+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
+Service: Knowledge creation from Case Info,Generation,Auto,XGen 22B,XGen 22B (1228),Salesforce,3.1875,3.25,3.0,3.1875,3.15625
+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
+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
+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
+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
+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
+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
+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
+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
+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
+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
+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
+Sales: Email Summary,Summary,Auto,Gemini  Pro 1,Gemini  Pro 1,Google,3.377551020408163,3.3877551020408165,3.193877551020408,3.4285714285714284,3.346938775510204
+Sales: Email Summary,Summary,Auto,GPT4-o,GPT4-o,OpenAI,3.8877551020408165,3.9693877551020407,3.86734693877551,3.8979591836734695,3.9056122448979593
+Sales: Email Summary,Summary,Auto,AI21 Jamba-Instruct,AI21 (jamba-instruct-preview),AI21,3.5714285714285716,3.8979591836734695,3.9183673469387754,3.663265306122449,3.7627551020408165
+Sales: Email Summary,Summary,Auto,SF-TextBase 70B,TextBase-70B (Llama FT),Salesforce,3.663265306122449,3.979591836734694,3.9591836734693877,3.836734693877551,3.8596938775510203
+Sales: Email Summary,Summary,Auto,Claude 3 Haiku,Claude 3 Haiku,Anthropic,3.857142857142857,3.989795918367347,3.9183673469387754,3.7653061224489797,3.88265306122449
+Sales: Email Summary,Summary,Auto,Cohere Command R+,cohere.cmd-R+,Cohere AI,3.7244897959183674,3.8979591836734695,3.795918367346939,3.7551020408163267,3.793367346938776
+Sales: Email Summary,Summary,Auto,XGen 22B,XGen 22B (1228),Salesforce,3.2783505154639174,2.979381443298969,3.142857142857143,2.8969072164948453,3.0743740795287184
+Sales: Email Summary,Summary,Auto,SF-TextBase 7B,TextBase-7B (Mistral FT),Salesforce,3.357142857142857,3.6020408163265305,3.663265306122449,3.510204081632653,3.5331632653061225
+Sales: Email Summary,Summary,Auto,LLaMA 3 70B,Meta-Llama-3-70B-Instruct,Meta,3.642857142857143,3.9693877551020407,3.9591836734693877,3.8979591836734695,3.86734693877551
+Sales: Email Summary,Summary,Auto,Cohere Command Text,cohere.command-text-v14,Cohere AI,3.154639175257732,2.9690721649484537,2.649484536082474,3.2783505154639174,3.0128865979381443
+Sales: Email Summary,Summary,Auto,GPT 4 Turbo,gpt-4-0613,OpenAI,3.826530612244898,4.0,3.9285714285714284,3.9591836734693877,3.928571428571429
+Sales: Email Summary,Summary,Auto,GPT 3.5 Turbo,gpt-3.5-turbo,OpenAI,3.7857142857142856,3.9591836734693877,3.9693877551020407,3.693877551020408,3.8520408163265305
+Sales: Email Summary,Summary,Auto,Claude 3 Opus,Claude 3 (Opus),Anthropic,3.826530612244898,4.0,3.9183673469387754,3.938775510204082,3.920918367346939
+Sales: Email Summary,Summary,Auto,Gemini  Pro 1.5,Gemini  Pro 1.5,Google,3.7653061224489797,3.9285714285714284,3.9183673469387754,3.663265306122449,3.8188775510204085
+Sales: Email Summary,Summary,Auto,LLaMA 3 8B,Meta-Llama-3-8B-Instruct,Meta,3.7346938775510203,4.0,3.9693877551020407,3.806122448979592,3.877551020408163
+Sales: Email Summary,Summary,Auto,Mistral 7B,Mistral-7B-Instruct-v0.1,Mistral,3.6288659793814433,3.948453608247423,3.7653061224489797,3.948453608247423,3.822769829581317
+Sales: Email Summary,Summary,Auto,Mixtral 8x7B,Mixtral-8x7B-v0.1,Mistral,3.5306122448979593,3.9591836734693877,3.8469387755102042,3.816326530612245,3.788265306122449
+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
+Service: Call Summary,Summary,Auto,Mistral 7B,Mistral-7B-Instruct-v0.1,Mistral,3.75,3.95,3.925,3.925,3.8875
+Service: Call Summary,Summary,Auto,Gemini  Pro 1,Gemini  Pro 1,Google,3.775,3.95,3.9,3.85,3.8687500000000004
+Service: Call Summary,Summary,Auto,GPT4-o,GPT4-o,OpenAI,3.925,3.95,3.95,3.925,3.9375
+Service: Call Summary,Summary,Auto,SF-TextBase 70B,TextBase-70B (Llama FT),Salesforce,3.975,3.95,3.95,3.95,3.9562500000000003
+Service: Call Summary,Summary,Auto,LLaMA 3 8B,Meta-Llama-3-8B-Instruct,Meta,3.9,3.95,3.95,3.95,3.9375000000000004
+Service: Call Summary,Summary,Auto,Claude 3 Haiku,Claude 3 Haiku,Anthropic,3.8,3.925,3.775,3.925,3.85625
+Service: Call Summary,Summary,Auto,Cohere Command R+,cohere.cmd-R+,Cohere AI,3.925,3.95,3.95,3.925,3.9375
+Service: Call Summary,Summary,Auto,Cohere Command Text,cohere.command-text-v14,Cohere AI,3.6,3.75,3.675,3.85,3.7187499999999996
+Service: Call Summary,Summary,Auto,SF-TextBase 7B,TextBase-7B (Mistral FT),Salesforce,3.95,4.0,3.975,3.875,3.95
+Service: Call Summary,Summary,Auto,XGen 22B,XGen 22B (1228),Salesforce,3.65,3.95,3.95,3.8,3.8375
+Service: Call Summary,Summary,Auto,LLaMA 3 70B,Meta-Llama-3-70B-Instruct,Meta,3.875,3.95,3.95,3.925,3.925
+Service: Call Summary,Summary,Auto,Claude 3 Opus,Claude 3 (Opus),Anthropic,3.95,4.0,4.0,4.0,3.9875
+Service: Call Summary,Summary,Auto,Mixtral 8x7B,Mixtral-8x7B-v0.1,Mistral,3.825,3.975,3.95,3.925,3.91875
+Service: Call Summary,Summary,Auto,Gemini  Pro 1.5,Gemini  Pro 1.5,Google,3.95,3.95,4.0,3.925,3.95625
+Service: Call Summary,Summary,Auto,GPT 3.5 Turbo,gpt-3.5-turbo,OpenAI,3.85,3.95,3.95,3.875,3.90625
+Service: Call Summary,Summary,Auto,GPT 4 Turbo,gpt-4-0613,OpenAI,3.9,3.95,3.95,3.925,3.93125
+Service: Call Summary,Summary,Auto,AI21 Jamba-Instruct,AI21 (jamba-instruct-preview),AI21,3.825,3.95,3.925,3.9,3.8999999999999995
+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
+Sales: Call Summary,Summary,Auto,Gemini  Pro 1,Gemini  Pro 1,Google,2.789473684210526,3.0526315789473686,2.473684210526316,2.8947368421052633,2.8026315789473686
+Sales: Call Summary,Summary,Auto,GPT4-o,GPT4-o,OpenAI,3.9473684210526314,3.9473684210526314,3.473684210526316,3.210526315789474,3.6447368421052633
+Sales: Call Summary,Summary,Auto,AI21 Jamba-Instruct,AI21 (jamba-instruct-preview),AI21,3.3684210526315788,3.3157894736842106,3.1052631578947367,3.0,3.1973684210526314
+Sales: Call Summary,Summary,Auto,SF-TextBase 70B,TextBase-70B (Llama FT),Salesforce,3.6842105263157894,3.3684210526315788,3.263157894736842,2.789473684210526,3.276315789473684
+Sales: Call Summary,Summary,Auto,Claude 3 Haiku,Claude 3 Haiku,Anthropic,3.6315789473684212,3.473684210526316,3.3157894736842106,3.0526315789473686,3.368421052631579
+Sales: Call Summary,Summary,Auto,Cohere Command R+,cohere.cmd-R+,Cohere AI,3.789473684210526,3.6842105263157894,2.8421052631578947,3.0526315789473686,3.3421052631578942
+Sales: Call Summary,Summary,Auto,XGen 22B,XGen 22B (1228),Salesforce,2.789473684210526,2.736842105263158,2.5789473684210527,2.6315789473684212,2.6842105263157894
+Sales: Call Summary,Summary,Auto,SF-TextBase 7B,TextBase-7B (Mistral FT),Salesforce,3.210526315789474,3.1052631578947367,3.210526315789474,2.6315789473684212,3.0394736842105265
+Sales: Call Summary,Summary,Auto,LLaMA 3 70B,Meta-Llama-3-70B-Instruct,Meta,3.4210526315789473,3.263157894736842,3.473684210526316,2.8421052631578947,3.2499999999999996
+Sales: Call Summary,Summary,Auto,Cohere Command Text,cohere.command-text-v14,Cohere AI,1.7894736842105263,2.473684210526316,2.1052631578947367,2.3157894736842106,2.1710526315789473
+Sales: Call Summary,Summary,Auto,GPT 4 Turbo,gpt-4-0613,OpenAI,4.0,3.736842105263158,3.3157894736842106,3.0,3.513157894736842
+Sales: Call Summary,Summary,Auto,GPT 3.5 Turbo,gpt-3.5-turbo,OpenAI,3.8421052631578947,3.3684210526315788,3.8421052631578947,2.888888888888889,3.485380116959064
+Sales: Call Summary,Summary,Auto,Claude 3 Opus,Claude 3 (Opus),Anthropic,3.9473684210526314,3.8947368421052633,3.5789473684210527,3.0,3.605263157894737
+Sales: Call Summary,Summary,Auto,Gemini  Pro 1.5,Gemini  Pro 1.5,Google,3.736842105263158,3.6315789473684212,3.0526315789473686,3.0526315789473686,3.3684210526315788
+Sales: Call Summary,Summary,Auto,LLaMA 3 8B,Meta-Llama-3-8B-Instruct,Meta,2.9473684210526314,3.1578947368421053,2.6315789473684212,2.8947368421052633,2.9078947368421053
+Sales: Call Summary,Summary,Auto,Mistral 7B,Mistral-7B-Instruct-v0.1,Mistral,3.5555555555555554,3.4210526315789473,3.210526315789474,3.1052631578947367,3.323099415204678
+Sales: Call Summary,Summary,Auto,Mixtral 8x7B,Mixtral-8x7B-v0.1,Mistral,3.263157894736842,3.263157894736842,2.6315789473684212,3.0,3.0394736842105265
+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
+Service: Live Chat Insights,Summary,Auto,Gemini  Pro 1,Gemini  Pro 1,Google,3.76953125,3.97265625,3.87109375,3.8359375,3.8623046875
+Service: Live Chat Insights,Summary,Auto,GPT4-o,GPT4-o,OpenAI,3.91015625,4.0,3.9765625,3.91796875,3.951171875
+Service: Live Chat Insights,Summary,Auto,AI21 Jamba-Instruct,AI21 (jamba-instruct-preview),AI21,3.859375,3.97265625,3.8515625,3.8203125,3.8759765625
+Service: Live Chat Insights,Summary,Auto,SF-TextBase 70B,TextBase-70B (Llama FT),Salesforce,3.85546875,3.99609375,3.94140625,3.96875,3.9404296875
+Service: Live Chat Insights,Summary,Auto,Claude 3 Haiku,Claude 3 Haiku,Anthropic,3.8828125,4.0,3.796875,3.98046875,3.9150390625
+Service: Live Chat Insights,Summary,Auto,Cohere Command R+,cohere.cmd-R+,Cohere AI,3.87109375,3.984375,3.8671875,3.9375,3.9150390625
+Service: Live Chat Insights,Summary,Auto,XGen 22B,XGen 22B (1228),Salesforce,3.74609375,3.9765625,3.921875,3.4375,3.7705078125
+Service: Live Chat Insights,Summary,Auto,SF-TextBase 7B,TextBase-7B (Mistral FT),Salesforce,3.71484375,3.97265625,3.90625,3.6875,3.8203125
+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
+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
+Service: Live Chat Insights,Summary,Auto,GPT 4 Turbo,gpt-4-0613,OpenAI,3.91015625,3.9921875,3.93359375,3.9375,3.943359375
+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
+Service: Live Chat Insights,Summary,Auto,Claude 3 Opus,Claude 3 (Opus),Anthropic,3.9140625,4.0,3.82421875,3.98046875,3.9296875
+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
+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
+Service: Live Chat Insights,Summary,Auto,Mistral 7B,Mistral-7B-Instruct-v0.1,Mistral,3.8125,3.9921875,3.7734375,3.875,3.86328125
+Service: Live Chat Insights,Summary,Auto,Mixtral 8x7B,Mixtral-8x7B-v0.1,Mistral,3.8046875,3.96875,3.6015625,3.953125,3.83203125
+Service: Live Chat Insights,Summary,Auto,SF-TextSum,Summarization model 7B for Service (Mistral FT),Salesforce,3.80859375,3.9765625,3.92578125,3.7734375,3.87109375
+Service: Knowledge creation from Case Info,Generation,Manual,GPT 3.5 Turbo,gpt-3.5-turbo,OpenAI,3.3125,3.1250000000000004,3.15625,3.0208333333333335,3.1536458333333335
+Service: Knowledge creation from Case Info,Generation,Manual,GPT 4 Turbo,gpt-4-0613,OpenAI,3.2604166666666665,3.078125,3.3125,3.375,3.2565104166666665
+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
+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
+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
+Service: Knowledge creation from Case Info,Generation,Manual,XGen 22B,XGen 22B (1228),Salesforce,2.880208333333333,2.921875,3.0416666666666665,2.84375,2.921875
+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
+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
+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
+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
+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
+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
+Service: Knowledge creation from Case Info,Generation,Manual,GPT4-o,GPT4-o,OpenAI,3.3020833333333335,3.296875,3.171875,3.380208333333334,3.287760416666667
+Service: Knowledge creation from Case Info,Generation,Manual,SF-TextSum,Summarization model 7B for Service (Mistral FT),Salesforce,3.1510416666666665,2.9114583333333335,3.046875,2.96875,3.01953125
+Service: Knowledge creation from Case Info,Generation,Manual,SF-TextBase 70B,TextBase-70B (Llama FT),Salesforce,3.208333333333333,3.057291666666667,3.114583333333333,3.0000000000000004,3.095052083333333
 Service: Reply Recommendations,Generation,Manual,GPT 3.5 Turbo,gpt-3.5-turbo,OpenAI,3.0935185185185183,3.22037037037037,3.443518518518519,3.0453703703703705,3.2006944444444443
 Service: Reply Recommendations,Generation,Manual,GPT 4 Turbo,gpt-4-0613,OpenAI,3.525925925925926,3.3203703703703704,3.5129629629629635,3.52962962962963,3.4722222222222228
 Service: Reply Recommendations,Generation,Manual,Mixtral 8x7B,Mixtral-8x7B-v0.1,Mistral,3.015740740740741,3.0203703703703706,3.2111111111111112,2.837037037037037,3.021064814814815
@@ -208,28 +223,14 @@ Service: Reply Recommendations,Generation,Manual,Cohere Command Text,cohere.comm
 Service: Reply Recommendations,Generation,Manual,Gemini  Pro 1,Gemini  Pro 1,Google,2.851851851851852,2.761111111111111,2.2962962962962963,3.2379629629629627,2.7868055555555555
 Service: Reply Recommendations,Generation,Manual,LLaMA 3 70B,Meta-Llama-3-70B-Instruct,Meta,3.2796296296296297,3.2101851851851846,3.230555555555556,3.021296296296297,3.185416666666667
 Service: Reply Recommendations,Generation,Manual,LLaMA 3 8B,Meta-Llama-3-8B-Instruct,Meta,3.087037037037037,3.1166666666666667,3.249074074074074,2.888888888888889,3.0854166666666667
-Sales: Email Generation,Generation,Manual,GPT 3.5 Turbo,gpt-3.5-turbo,OpenAI,3.037735849056604,2.515723270440251,3.204402515723271,2.811320754716981,2.892295597484277
-Sales: Email Generation,Generation,Manual,GPT 4 Turbo,gpt-4-0613,OpenAI,3.2169811320754724,3.067610062893082,2.9088050314465415,3.248427672955976,3.1104559748427683
-Sales: Email Generation,Generation,Manual,Mixtral 8x7B,Mixtral-8x7B-v0.1,Mistral,3.1163522012578615,2.5345911949685536,2.9937106918238996,2.8899371069182394,2.8836477987421385
-Sales: Email Generation,Generation,Manual,AI21 Jamba-Instruct,AI21 (jamba-instruct-preview),AI21,3.0817610062893084,2.1132075471698113,2.7106918238993716,2.880503144654088,2.6965408805031448
-Sales: Email Generation,Generation,Manual,Mistral 7B,Mistral-7B-Instruct-v0.1,Mistral,2.9056603773584904,1.9182389937106918,2.883647798742138,2.60062893081761,2.5770440251572326
-Sales: Email Generation,Generation,Manual,XGen 22B,XGen 22B (1228),Salesforce,3.0031446540880506,2.1540880503144657,2.90880503144654,2.7044025157232707,2.6926100628930816
-Sales: Email Generation,Generation,Manual,Claude 3 Haiku,Claude 3 Haiku,Anthropic,3.0880503144654092,2.5283018867924527,3.0220125786163528,3.1635220125786163,2.9504716981132075
-Sales: Email Generation,Generation,Manual,Cohere Command Text,cohere.command-text-v14,Cohere AI,2.878930817610063,2.1635220125786163,2.858490566037736,2.7106918238993707,2.6529088050314464
-Sales: Email Generation,Generation,Manual,Gemini  Pro 1,Gemini  Pro 1,Google,3.1163522012578633,2.5911949685534585,2.757861635220126,2.9622641509433967,2.8569182389937113
-Sales: Email Generation,Generation,Manual,LLaMA 3 70B,Meta-Llama-3-70B-Instruct,Meta,3.0943396226415096,3.0062893081761004,3.2232704402515724,3.1415094339622636,3.116352201257861
-Sales: Email Generation,Generation,Manual,LLaMA 3 8B,Meta-Llama-3-8B-Instruct,Meta,3.213836477987421,3.0062893081761013,2.9937106918238996,3.0534591194968557,3.066823899371069
-Sales: Email Generation,Generation,Manual,SF-TextBase 7B,TextBase-7B (Mistral FT),Salesforce,3.169871794871795,2.4839743589743586,2.9519230769230775,2.7660256410256405,2.8429487179487176
-Sales: Email Generation,Generation,Manual,GPT4-o,GPT4-o,OpenAI,3.286163522012579,3.2389937106918247,3.2955974842767293,3.2421383647798745,3.265723270440252
-Sales: Email Generation,Generation,Manual,SF-TextBase 70B,TextBase-70B (Llama FT),Salesforce,3.1729559748427683,3.09119496855346,3.172955974842768,3.2012578616352205,3.159591194968554
 Service: Call Summary,Summary,Manual,GPT 3.5 Turbo,gpt-3.5-turbo,OpenAI,3.5694444444444446,3.3541666666666665,3.7948717948717947,3.4193548387096775,3.534459436173146
-Service: Call Summary,Summary,Manual,GPT 4 Turbo,gpt-4-0613,OpenAI,3.4743589743589745,3.4871794871794872,3.5128205128205128,3.4615384615384617,3.483974358974359
-Service: Call Summary,Summary,Manual,Mixtral 8x7B,Mixtral-8x7B-v0.1,Mistral,2.725,2.6875,2.782051282051282,2.675,2.717387820512821
-Service: Call Summary,Summary,Manual,AI21 Jamba-Instruct,AI21 (jamba-instruct-preview),AI21,2.58974358974359,2.7125,2.7435897435897436,2.6538461538461537,2.674919871794872
-Service: Call Summary,Summary,Manual,Mistral 7B,Mistral-7B-Instruct-v0.1,Mistral,2.75,2.775,2.8684210526315788,2.7,2.773355263157895
-Service: Call Summary,Summary,Manual,XGen 22B,XGen 22B (1228),Salesforce,2.7,2.8125,2.9615384615384617,2.7,2.7935096153846155
-Service: Call Summary,Summary,Manual,Claude 3 Haiku,Claude 3 Haiku,Anthropic,2.525,2.6,2.7222222222222223,2.5875,2.6086805555555554
-Service: Call Summary,Summary,Manual,Cohere Command Text,cohere.command-text-v14,Cohere AI,2.35,2.4125,2.527027027027027,2.375,2.416131756756757
+Service: Call Summary,Summary,Manual,GPT 4 Turbo,gpt-4-0613,OpenAI,3.4805194805194803,3.4935064935064934,3.5194805194805197,3.4675324675324677,3.4902597402597406
+Service: Call Summary,Summary,Manual,Mixtral 8x7B,Mixtral-8x7B-v0.1,Mistral,2.721518987341772,2.6962025316455698,2.779220779220779,2.6835443037974684,2.7201216505013974
+Service: Call Summary,Summary,Manual,AI21 Jamba-Instruct,AI21 (jamba-instruct-preview),AI21,2.5844155844155843,2.7088607594936707,2.74025974025974,2.6493506493506493,2.6707216833799112
+Service: Call Summary,Summary,Manual,Mistral 7B,Mistral-7B-Instruct-v0.1,Mistral,2.7468354430379747,2.7721518987341773,2.8666666666666667,2.6962025316455698,2.770464135021097
+Service: Call Summary,Summary,Manual,XGen 22B,XGen 22B (1228),Salesforce,2.6962025316455698,2.810126582278481,2.961038961038961,2.6962025316455698,2.7908926516521455
+Service: Call Summary,Summary,Manual,Claude 3 Haiku,Claude 3 Haiku,Anthropic,2.518987341772152,2.5949367088607596,2.7183098591549295,2.5822784810126582,2.603628097700125
+Service: Call Summary,Summary,Manual,Cohere Command Text,cohere.command-text-v14,Cohere AI,2.3417721518987342,2.4050632911392404,2.5205479452054793,2.367088607594937,2.4086179989595977
 Service: Call Summary,Summary,Manual,Gemini  Pro 1,Gemini  Pro 1,Google,2.2051282051282053,2.3684210526315788,2.442857142857143,2.217948717948718,2.3085887796414113
 Service: Call Summary,Summary,Manual,LLaMA 3 70B,Meta-Llama-3-70B-Instruct,Meta,2.7435897435897436,2.769230769230769,2.9285714285714284,2.6666666666666665,2.777014652014652
 Service: Call Summary,Summary,Manual,LLaMA 3 8B,Meta-Llama-3-8B-Instruct,Meta,2.5135135135135136,2.6052631578947367,2.6818181818181817,2.4871794871794872,2.57194358510148
@@ -248,3 +249,32 @@ Service: Live Chat Insights,Summary,Manual,SF-TextSum,Summarization model 7B for
 Service: Live Chat Insights,Summary,Manual,SF-TextBase 7B,TextBase-7B (Mistral FT),Salesforce,3.56,3.64,3.68,3.66,3.6350000000000002
 Service: Live Chat Insights,Summary,Manual,GPT4-o,GPT4-o,OpenAI,3.78,3.78,3.78,3.78,3.78
 Service: Live Chat Insights,Summary,Manual,SF-TextBase 70B,TextBase-70B (Llama FT),Salesforce,3.72,3.74,3.6,3.72,3.6950000000000003
+Sales: Call Summary,Summary,Manual,GPT 3.5 Turbo,gpt-3.5-turbo,OpenAI,2.648148148148148,2.6759259259259256,2.8796296296296293,2.4629629629629632,2.6666666666666665
+Sales: Call Summary,Summary,Manual,GPT 4 Turbo,gpt-4-0613,OpenAI,3.185185185185185,3.037037037037037,3.0925925925925926,2.9444444444444446,3.064814814814815
+Sales: Call Summary,Summary,Manual,Mixtral 8x7B,Mixtral-8x7B-v0.1,Mistral,2.5,2.5370370370370368,2.509259259259259,2.5462962962962963,2.523148148148148
+Sales: Call Summary,Summary,Manual,AI21 Jamba-Instruct,AI21 (jamba-instruct-preview),AI21,2.351851851851852,2.5555555555555554,2.7407407407407405,2.3796296296296293,2.506944444444444
+Sales: Call Summary,Summary,Manual,Mistral 7B,Mistral-7B-Instruct-v0.1,Mistral,2.4444444444444446,2.6203703703703707,2.7685185185185186,2.3333333333333335,2.541666666666667
+Sales: Call Summary,Summary,Manual,XGen 22B,XGen 22B (1228),Salesforce,2.4074074074074074,2.490740740740741,2.6574074074074074,2.240740740740741,2.4490740740740744
+Sales: Call Summary,Summary,Manual,Claude 3 Haiku,Claude 3 Haiku,Anthropic,2.898148148148148,2.75,2.935185185185185,2.5555555555555554,2.784722222222222
+Sales: Call Summary,Summary,Manual,Cohere Command Text,cohere.command-text-v14,Cohere AI,2.074074074074074,2.2314814814814814,2.3703703703703702,1.9351851851851853,2.1527777777777777
+Sales: Call Summary,Summary,Manual,Gemini  Pro 1,Gemini  Pro 1,Google,2.7777777777777777,2.6666666666666665,2.7870370370370368,2.8333333333333335,2.7662037037037037
+Sales: Call Summary,Summary,Manual,LLaMA 3 70B,Meta-Llama-3-70B-Instruct,Meta,2.657407407407407,2.6018518518518516,3.046296296296296,2.3055555555555554,2.6527777777777772
+Sales: Call Summary,Summary,Manual,LLaMA 3 8B,Meta-Llama-3-8B-Instruct,Meta,2.4814814814814814,2.685185185185185,2.768518518518518,2.2129629629629632,2.5370370370370368
+Sales: Call Summary,Summary,Manual,SF-TextBase 7B,TextBase-7B (Mistral FT),Salesforce,2.490740740740741,2.5462962962962963,2.8240740740740744,2.1203703703703702,2.4953703703703702
+Sales: Call Summary,Summary,Manual,GPT4-o,GPT4-o,OpenAI,3.1568627450980395,3.156862745098039,2.9411764705882355,3.098039215686274,3.088235294117647
+Sales: Call Summary,Summary,Manual,SF-TextSum,Summarization model 7B for Service (Mistral FT),Salesforce,2.1666666666666665,2.401960784313726,2.656862745098039,2.068627450980392,2.3235294117647056
+Sales: Call Summary,Summary,Manual,SF-TextBase 70B,TextBase-70B (Llama FT),Salesforce,2.601851851851852,2.8981481481481484,3.009259259259259,2.342592592592593,2.7129629629629632
+Sales: Email Generation,Generation,Manual,GPT 3.5 Turbo,gpt-3.5-turbo,OpenAI,3.2113207547169815,2.6641509433962263,3.2446540880503143,2.864150943396226,2.996069182389937
+Sales: Email Generation,Generation,Manual,GPT 4 Turbo,gpt-4-0613,OpenAI,3.3616352201257866,3.0377358490566038,3.053459119496856,3.1540880503144657,3.151729559748428
+Sales: Email Generation,Generation,Manual,Mixtral 8x7B,Mixtral-8x7B-v0.1,Mistral,3.2317610062893083,2.4915094339622645,3.0503144654088055,2.949685534591195,2.9308176100628933
+Sales: Email Generation,Generation,Manual,AI21 Jamba-Instruct,AI21 (jamba-instruct-preview),AI21,3.162758310871518,2.105031446540881,2.833333333333333,2.986477987421384,2.771900269541779
+Sales: Email Generation,Generation,Manual,Mistral 7B,Mistral-7B-Instruct-v0.1,Mistral,3.0338274932614566,1.855031446540881,2.9367924528301894,2.598787061994609,2.606109613656784
+Sales: Email Generation,Generation,Manual,XGen 22B,XGen 22B (1228),Salesforce,2.982704402515723,2.037106918238994,2.9817610062893083,2.619811320754717,2.6553459119496856
+Sales: Email Generation,Generation,Manual,Claude 3 Haiku,Claude 3 Haiku,Anthropic,3.111635220125786,2.5229559748427675,3.061006289308176,3.0185534591194974,2.9285377358490567
+Sales: Email Generation,Generation,Manual,Cohere Command Text,cohere.command-text-v14,Cohere AI,2.912309074573226,2.0734950584007192,2.843800539083558,2.5796046720575014,2.602302336028751
+Sales: Email Generation,Generation,Manual,Gemini  Pro 1,Gemini  Pro 1,Google,3.1317610062893073,2.430503144654088,2.7952830188679245,2.947798742138365,2.826336477987421
+Sales: Email Generation,Generation,Manual,LLaMA 3 70B,Meta-Llama-3-70B-Instruct,Meta,3.0899371069182386,2.9817610062893087,3.155974842767295,3.08993710691824,3.0794025157232707
+Sales: Email Generation,Generation,Manual,LLaMA 3 8B,Meta-Llama-3-8B-Instruct,Meta,3.251572327044026,3.0471698113207544,3.0764150943396227,3.133647798742139,3.1272012578616355
+Sales: Email Generation,Generation,Manual,SF-TextBase 7B,TextBase-7B (Mistral FT),Salesforce,3.155974842767295,2.359119496855346,3.0169811320754722,2.7386792452830186,2.8176886792452827
+Sales: Email Generation,Generation,Manual,GPT4-o,GPT4-o,OpenAI,3.3072327044025167,3.1880503144654093,3.3264150943396222,3.219496855345912,3.2602987421383656
+Sales: Email Generation,Generation,Manual,SF-TextBase 70B,TextBase-70B (Llama FT),Salesforce,3.0911949685534585,3.006603773584905,3.1801886792452825,3.203459119496855,3.1203616352201253

src/about.py

@@ -33,7 +33,20 @@ INTRODUCTION_TEXT = """
 
 # Which evaluations are you running? how can people reproduce what you have?
 LLM_BENCHMARKS_TEXT = """
-
+1) GPT-4T was used except for some accuracy use cases with atypically long input tokens.
+2) Hyperparameters were optimized for a subset of models evaluated (platform models?) Were parameters optimized as well?
+3) Latency reflects the mean latency over a single time range on a high-speed internet connection; response times for external APIs may vary over time and be impacted by internet speed, location, etc.
+3) Latency reflects the time to receive the entire completion.
+4) Some external APIs were direct to the LLM provider (OpenAI, Google, AI21), while others were provided through Amazon Bedrock (Cohere, Anthropic)
+5) LLM annotations (manual/human evaluations) were performed under a variety of settings that did not necessarily control for ordering effects
+6) All tests on open source models were performed on original models (correct?); custom fine-tuning may impact performance in trust / safety / toxicity / bias / etc.
+7) For the tests on latency, the inputs were *approximately* 500 / 3000 tokens. A short prompt was added and different models tokenize differently.
+8) Costs for all external APIs were based on the standard pricing of the provider (note that the pricing of cohere/anthropic via Bedrock is the same as directly through Cohere/Anthropic apis).
+9) Something about limitations of LLM judges (despite correlation with human annotators)
+10) Task-specific model variants were not used from the external providers (command-r is sort of retrieval specific, but this was not one of the use cases)
+11) Maybe something about the tasks being primarily summarization / generation
+12) CRM T&S is done by perturbing words: 1) for gender bias, we perturb person names and pronouns to opposite gender. 2) for entity bias, we perturb company names to its competitors in the same sector
+13) Cost per request for self-hosted models assume a minimal frequency of calling the model, since the costs are per hour. All latencies / cost assume a single user at a time.
 """
 
 EVALUATION_QUEUE_TEXT = """
@@ -68,4 +81,11 @@ If everything is done, check you can launch the EleutherAIHarness on your model
 
 CITATION_BUTTON_LABEL = "Copy the following snippet to cite these results"
 CITATION_BUTTON_TEXT = r"""
+@misc{crm-llm-leaderboard,
+  author = {Salesforce AI},
+  title = {LLM Leaderboard for CRM},
+  year = {2024},
+  publisher = {Salesforce AI},
+  howpublished = "\url{https://https://huggingface.co/spaces/Salesforce/crm_llm_leaderboard}"
+}
 """

src/display/utils.py

@@ -33,11 +33,11 @@ auto_eval_column_dict.append(
     ["model_provider", ColumnContent, ColumnContent("LLM Provider", "markdown", True, never_hidden=True)]
 )
 auto_eval_column_dict.append(["use_case_name", ColumnContent, ColumnContent("Use Case Name", "markdown", True)])
-auto_eval_column_dict.append(["use_case_type", ColumnContent, ColumnContent("Use Case Type", "markdown", True)])
+auto_eval_column_dict.append(["use_case_type", ColumnContent, ColumnContent("Use Case Type", "markdown", False)])
 
-auto_eval_column_dict.append(["accuracy_method", ColumnContent, ColumnContent("Accuracy Method", "markdown", True)])
+auto_eval_column_dict.append(["accuracy_method", ColumnContent, ColumnContent("Accuracy Method", "markdown", False)])
 # Accuracy metrics
-auto_eval_column_dict.append(["accuracy_metric_average", ColumnContent, ColumnContent("Average", "markdown", True)])
+auto_eval_column_dict.append(["accuracy_metric_average", ColumnContent, ColumnContent("Accuracy", "markdown", True)])
 auto_eval_column_dict.append(
     [
         "accuracy_metric_instruction_following",
@@ -56,7 +56,6 @@ auto_eval_column_dict.append(
 )
 # Speed (Latency) metrics
 
-
 # Cost metrics
 
 # Trust & Safety metrics