Text Generation
Transformers
GGUF
English
mistral
text-generation-inference
Xin Liu commited on
Commit
487bf15
1 Parent(s): 15861bd

Signed-off-by: Xin Liu <sam@secondstate.io>

Files changed (1) hide show
  1. README.md +78 -9
README.md CHANGED
@@ -1,11 +1,80 @@
1
- # dolphin-2.6-mistral-7B
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
2
 
3
- ## Prompt Template: ChatML
 
 
 
 
 
 
4
 
5
- ```bash
6
- <|im_start|>system
7
- {system_message}<|im_end|>
8
- <|im_start|>user
9
- {prompt}<|im_end|>
10
- <|im_start|>assistant
11
- ```
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ base_model: cognitivecomputations/dolphin-2.6-mistral-7b
3
+ inference: false
4
+ library_name: transformers
5
+ license: apache-2.0
6
+ model_creator: Cognitive Computations
7
+ model_name: Dolphin 2.6 Mistral
8
+ datasets:
9
+ - ehartford/dolphin
10
+ - jondurbin/airoboros-2.2.1
11
+ - ehartford/dolphin-coder
12
+ - teknium/openhermes
13
+ - ise-uiuc/Magicoder-OSS-Instruct-75K
14
+ - ise-uiuc/Magicoder-Evol-Instruct-110K
15
+ - LDJnr/Capybara
16
+ language:
17
+ - en
18
+ quantized_by: Second State Inc.
19
+ ---
20
 
21
+ <!-- header start -->
22
+ <!-- 200823 -->
23
+ <div style="width: auto; margin-left: auto; margin-right: auto">
24
+ <img src="https://github.com/second-state/LlamaEdge/raw/dev/assets/logo.svg" style="width: 100%; min-width: 400px; display: block; margin: auto;">
25
+ </div>
26
+ <hr style="margin-top: 1.0em; margin-bottom: 1.0em;">
27
+ <!-- header end -->
28
 
29
+ # Dolphin-2.6-Mistral-7B-GGUF
30
+
31
+ ## Original Model
32
+
33
+ [cognitivecomputations/dolphin-2.6-mistral-7b](https://huggingface.co/cognitivecomputations/dolphin-2.6-mistral-7b)
34
+
35
+ ## Run with LlamaEdge
36
+
37
+ - LlamaEdge version: [v0.2.4](https://github.com/second-state/LlamaEdge/releases/tag/0.2.4)
38
+
39
+ - Prompt template
40
+
41
+ - Prompt type: `chatml`
42
+
43
+ - Prompt string
44
+
45
+ ```text
46
+ <|im_start|>system
47
+ {system_message}<|im_end|>
48
+ <|im_start|>user
49
+ {prompt}<|im_end|>
50
+ <|im_start|>assistant
51
+ ```
52
+
53
+ - Run as LlamaEdge service
54
+
55
+ ```bash
56
+ wasmedge --dir .:. --nn-preload default:GGML:AUTO:dolphin-2.6-mistral-7b-Q5_K_M.gguf llama-api-server.wasm -p chatml
57
+ ```
58
+
59
+ - Run as LlamaEdge command app
60
+
61
+ ```bash
62
+ wasmedge --dir .:. --nn-preload default:GGML:AUTO:dolphin-2.6-mistral-7b-Q5_K_M.gguf llama-chat.wasm -p chatml
63
+ ```
64
+
65
+ ## Quantized GGUF Models
66
+
67
+ | Name | Quant method | Bits | Size | Use case |
68
+ | ---- | ---- | ---- | ---- | ----- |
69
+ | [dolphin-2.6-mistral-7b-Q2_K.gguf](https://huggingface.co/second-state/dolphin-2.6-mistral-7B-GGUF/blob/main/dolphin-2.6-mistral-7b-Q2_K.gguf) | Q2_K | 2 | 3.08 GB| smallest, significant quality loss - not recommended for most purposes |
70
+ | [dolphin-2.6-mistral-7b-Q3_K_L.gguf](https://huggingface.co/second-state/dolphin-2.6-mistral-7B-GGUF/blob/main/dolphin-2.6-mistral-7b-Q3_K_L.gguf) | Q3_K_L | 3 | 3.2 GB| small, substantial quality loss |
71
+ | [dolphin-2.6-mistral-7b-Q3_K_M.gguf](https://huggingface.co/second-state/dolphin-2.6-mistral-7B-GGUF/blob/main/dolphin-2.6-mistral-7b-Q3_K_M.gguf) | Q3_K_M | 3 | 3.52 GB| very small, high quality loss |
72
+ | [dolphin-2.6-mistral-7b-Q3_K_S.gguf](https://huggingface.co/second-state/dolphin-2.6-mistral-7B-GGUF/blob/main/dolphin-2.6-mistral-7b-Q3_K_S.gguf) | Q3_K_S | 3 | 3.16 GB| very small, high quality loss |
73
+ | [dolphin-2.6-mistral-7b-Q4_0.gguf](https://huggingface.co/second-state/dolphin-2.6-mistral-7B-GGUF/blob/main/dolphin-2.6-mistral-7b-Q4_0.gguf) | Q4_0 | 4 | 4.11 GB| legacy; small, very high quality loss - prefer using Q3_K_M |
74
+ | [dolphin-2.6-mistral-7b-Q4_K_M.gguf](https://huggingface.co/second-state/dolphin-2.6-mistral-7B-GGUF/blob/main/dolphin-2.6-mistral-7b-Q4_K_M.gguf) | Q4_K_M | 4 | 4.37 GB| medium, balanced quality - recommended |
75
+ | [dolphin-2.6-mistral-7b-Q4_K_S.gguf](https://huggingface.co/second-state/dolphin-2.6-mistral-7B-GGUF/blob/main/dolphin-2.6-mistral-7b-Q4_K_S.gguf) | Q4_K_S | 4 | 4.14 GB| small, greater quality loss |
76
+ | [dolphin-2.6-mistral-7b-Q5_0.gguf](https://huggingface.co/second-state/dolphin-2.6-mistral-7B-GGUF/blob/main/dolphin-2.6-mistral-7b-Q5_0.gguf) | Q5_0 | 5 | 5 GB| legacy; medium, balanced quality - prefer using Q4_K_M |
77
+ | [dolphin-2.6-mistral-7b-Q5_K_M.gguf](https://huggingface.co/second-state/dolphin-2.6-mistral-7B-GGUF/blob/main/dolphin-2.6-mistral-7b-Q5_K_M.gguf) | Q5_K_M | 5 | 5.13 GB| large, very low quality loss - recommended |
78
+ | [dolphin-2.6-mistral-7b-Q5_K_S.gguf](https://huggingface.co/second-state/dolphin-2.6-mistral-7B-GGUF/blob/main/dolphin-2.6-mistral-7b-Q5_K_S.gguf) | Q5_K_S | 5 | 5 GB| large, low quality loss - recommended |
79
+ | [dolphin-2.6-mistral-7b-Q6_K.gguf](https://huggingface.co/second-state/dolphin-2.6-mistral-7B-GGUF/blob/main/dolphin-2.6-mistral-7b-Q6_K.gguf) | Q6_K | 6 | 5.94 GB| very large, extremely low quality loss |
80
+ | [dolphin-2.6-mistral-7b-Q8_0.gguf](https://huggingface.co/second-state/dolphin-2.6-mistral-7B-GGUF/blob/main/dolphin-2.6-mistral-7b-Q8_0.gguf) | Q8_0 | 8 | 7.7 GB| very large, extremely low quality loss - not recommended |