Hugging Face's logo

second-state
/
Nous-Hermes-2-Mixtral-8x7B-DPO-GGUF

Text Generation
Transformers
GGUF
English
mixtral
Mixtral
instruct
finetune
chatml
DPO
RLHF
gpt4
synthetic data
distillation
conversational
Model card Files Community
Xin Liu commited on
Commit
bf47058
•
1 Parent(s): ab9305d

Update

Browse files

Signed-off-by: Xin Liu <[email protected]>

Files changed (1) hide show
  1. README.md +44 -21
README.md CHANGED
@@ -1,5 +1,26 @@
1
  ---
 
 
2
  license: apache-2.0
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
3
  ---
4
 
5
  <!-- header start -->
@@ -10,11 +31,11 @@ license: apache-2.0
10
  <hr style="margin-top: 1.0em; margin-bottom: 1.0em;">
11
  <!-- header end -->
12
 
13
- # OpenChat-3.5-0106-GGUF
14
 
15
  ## Original Model
16
 
17
- [openchat/openchat-3.5-0106](https://huggingface.co/openchat/openchat-3.5-0106)
18
 
19
  ## Run with LlamaEdge
20
 
@@ -22,41 +43,43 @@ license: apache-2.0
22
 
23
  - Prompt template
24
 
25
- - Prompt type: `openchat`
26
-
27
  - Prompt string
28
 
29
  ```text
30
- GPT4 User: {prompt}<|end_of_turn|>GPT4 Assistant:
 
 
 
 
31
  ```
32
 
33
- - Reverse prompt: `<|end_of_turn|>`
34
-
35
  - Run as LlamaEdge service
36
 
37
  ```bash
38
- wasmedge --dir .:. --nn-preload default:GGML:AUTO:openchat-3.5-0106-Q5_K_M.gguf llama-api-server.wasm -p openchat -r '<|end_of_turn|>'
39
  ```
40
 
41
  - Run as LlamaEdge command app
42
 
43
  ```bash
44
- wasmedge --dir .:. --nn-preload default:GGML:AUTO:openchat-3.5-0106-Q5_K_M.gguf llama-chat.wasm -p openchat -r '<|end_of_turn|>'
45
  ```
46
 
47
  ## Quantized GGUF Models
48
 
49
  | Name | Quant method | Bits | Size | Use case |
50
  | ---- | ---- | ---- | ---- | ----- |
51
- | [openchat-3.5-0106.Q2_K.gguf](https://huggingface.co/second-state/OpenChat-3.5-0106-GGUF/blob/main/openchat-3.5-0106-Q2_K.gguf) | Q2_K | 2 | 3.08 GB| smallest, significant quality loss - not recommended for most purposes |
52
- | [openchat-3.5-0106.Q3_K_L.gguf](https://huggingface.co/second-state/OpenChat-3.5-0106-GGUF/blob/main/openchat-3.5-0106-Q3_K_L.gguf) | Q3_K_L | 3 | 3.82 GB| small, substantial quality loss |
53
- | [openchat-3.5-0106.Q3_K_M.gguf](https://huggingface.co/second-state/OpenChat-3.5-0106-GGUF/blob/main/openchat-3.5-0106-Q3_K_M.gguf) | Q3_K_M | 3 | 3.52 GB| very small, high quality loss |
54
- | [openchat-3.5-0106.Q3_K_S.gguf](https://huggingface.co/second-state/OpenChat-3.5-0106-GGUF/blob/main/openchat-3.5-0106-Q3_K_S.gguf) | Q3_K_S | 3 | 3.16 GB| very small, high quality loss |
55
- | [openchat-3.5-0106.Q4_0.gguf](https://huggingface.co/second-state/OpenChat-3.5-0106-GGUF/blob/main/openchat-3.5-0106-Q4_0.gguf) | Q4_0 | 4 | 4.11 GB| legacy; small, very high quality loss - prefer using Q3_K_M |
56
- | [openchat-3.5-0106.Q4_K_M.gguf](https://huggingface.co/second-state/OpenChat-3.5-0106-GGUF/blob/main/openchat-3.5-0106-Q4_K_M.gguf) | Q4_K_M | 4 | 4.37 GB| medium, balanced quality - recommended |
57
- | [openchat-3.5-0106.Q4_K_S.gguf](https://huggingface.co/second-state/OpenChat-3.5-0106-GGUF/blob/main/openchat-3.5-0106-Q4_K_S.gguf) | Q4_K_S | 4 | 4.14 GB| small, greater quality loss |
58
- | [openchat-3.5-0106.Q5_0.gguf](https://huggingface.co/second-state/OpenChat-3.5-0106-GGUF/blob/main/openchat-3.5-0106-Q5_0.gguf) | Q5_0 | 5 | 5.00 GB| legacy; medium, balanced quality - prefer using Q4_K_M |
59
- | [openchat-3.5-0106.Q5_K_M.gguf](https://huggingface.co/second-state/OpenChat-3.5-0106-GGUF/blob/main/openchat-3.5-0106-Q5_K_M.gguf) | Q5_K_M | 5 | 5.13 GB| large, very low quality loss - recommended |
60
- | [openchat-3.5-0106.Q5_K_S.gguf](https://huggingface.co/second-state/OpenChat-3.5-0106-GGUF/blob/main/openchat-3.5-0106-Q5_K_S.gguf) | Q5_K_S | 5 | 5.00 GB| large, low quality loss - recommended |
61
- | [openchat-3.5-0106.Q6_K.gguf](https://huggingface.co/second-state/OpenChat-3.5-0106-GGUF/blob/main/openchat-3.5-0106-Q6_K.gguf) | Q6_K | 6 | 5.94 GB| very large, extremely low quality loss |
62
- | [openchat-3.5-0106.Q8_0.gguf](https://huggingface.co/second-state/OpenChat-3.5-0106-GGUF/blob/main/openchat-3.5-0106-Q8_0.gguf) | Q8_0 | 8 | 7.70 GB| very large, extremely low quality loss - not recommended |
 
1
  ---
2
+ base_model: NousResearch/Nous-Hermes-2-Mixtral-8x7B-DPO
3
+ inference: false
4
  license: apache-2.0
5
+ model-index:
6
+ - name: Nous-Hermes-2-Mixtral-8x7B-DPO
7
+ results: []
8
+ model_creator: NousResearch
9
+ model_name: Nous Hermes 2 Mixtral 8X7B DPO
10
+ model_type: mixtral
11
+ quantized_by: Second State Inc.
12
+ language:
13
+ - en
14
+ tags:
15
+ - Mixtral
16
+ - instruct
17
+ - finetune
18
+ - chatml
19
+ - DPO
20
+ - RLHF
21
+ - gpt4
22
+ - synthetic data
23
+ - distillation
24
  ---
25
 
26
  <!-- header start -->
 
31
  <hr style="margin-top: 1.0em; margin-bottom: 1.0em;">
32
  <!-- header end -->
33
 
34
+ # Nous-Hermes-2-Mixtral-8x7B-DPO-GGUF
35
 
36
  ## Original Model
37
 
38
+ [NousResearch/Nous-Hermes-2-Mixtral-8x7B-DPO](https://huggingface.co/NousResearch/Nous-Hermes-2-Mixtral-8x7B-DPO)
39
 
40
  ## Run with LlamaEdge
41
 
 
43
 
44
  - Prompt template
45
 
46
+ - Prompt type: `chatml`
47
+
48
  - Prompt string
49
 
50
  ```text
51
+ <|im_start|>system
52
+ {system_message}<|im_end|>
53
+ <|im_start|>user
54
+ {prompt}<|im_end|>
55
+ <|im_start|>assistant
56
  ```
57
 
 
 
58
  - Run as LlamaEdge service
59
 
60
  ```bash
61
+ wasmedge --dir .:. --nn-preload default:GGML:AUTO:Nous-Hermes-2-Mixtral-8x7B-DPO-Q5_K_M.gguf llama-api-server.wasm -p chatml
62
  ```
63
 
64
  - Run as LlamaEdge command app
65
 
66
  ```bash
67
+ wasmedge --dir .:. --nn-preload default:GGML:AUTO:Nous-Hermes-2-Mixtral-8x7B-DPO-Q5_K_M.gguf llama-chat.wasm -p chatml
68
  ```
69
 
70
  ## Quantized GGUF Models
71
 
72
  | Name | Quant method | Bits | Size | Use case |
73
  | ---- | ---- | ---- | ---- | ----- |
74
+ | [Nous-Hermes-2-Mixtral-8x7B-DPO-Q2_K.gguf](https://huggingface.co/second-state/Nous-Hermes-2-Mixtral-8x7B-DPO-GGUF/blob/main/Nous-Hermes-2-Mixtral-8x7B-DPO-Q2_K.gguf) | Q2_K | 2 | 17.3 GB| smallest, significant quality loss - not recommended for most purposes |
75
+ | [Nous-Hermes-2-Mixtral-8x7B-DPO-Q3_K_L.gguf](https://huggingface.co/second-state/Nous-Hermes-2-Mixtral-8x7B-DPO-GGUF/blob/main/Nous-Hermes-2-Mixtral-8x7B-DPO-Q3_K_L.gguf) | Q3_K_L | 3 | 24.2 GB| small, substantial quality loss |
76
+ | [Nous-Hermes-2-Mixtral-8x7B-DPO-Q3_K_M.gguf](https://huggingface.co/second-state/Nous-Hermes-2-Mixtral-8x7B-DPO-GGUF/blob/main/Nous-Hermes-2-Mixtral-8x7B-DPO-Q3_K_M.gguf) | Q3_K_M | 3 | 22.5 GB| very small, high quality loss |
77
+ | [Nous-Hermes-2-Mixtral-8x7B-DPO-Q3_K_S.gguf](https://huggingface.co/second-state/Nous-Hermes-2-Mixtral-8x7B-DPO-GGUF/blob/main/Nous-Hermes-2-Mixtral-8x7B-DPO-Q3_K_S.gguf) | Q3_K_S | 3 | 20.4 GB| very small, high quality loss |
78
+ | [Nous-Hermes-2-Mixtral-8x7B-DPO-Q4_0.gguf](https://huggingface.co/second-state/Nous-Hermes-2-Mixtral-8x7B-DPO-GGUF/blob/main/Nous-Hermes-2-Mixtral-8x7B-DPO-Q4_0.gguf) | Q4_0 | 4 | 26.4 GB| legacy; small, very high quality loss - prefer using Q3_K_M |
79
+ | [Nous-Hermes-2-Mixtral-8x7B-DPO-Q4_K_M.gguf](https://huggingface.co/second-state/Nous-Hermes-2-Mixtral-8x7B-DPO-GGUF/blob/main/Nous-Hermes-2-Mixtral-8x7B-DPO-Q4_K_M.gguf) | Q4_K_M | 4 | 28.4 GB| medium, balanced quality - recommended |
80
+ | [Nous-Hermes-2-Mixtral-8x7B-DPO-Q4_K_S.gguf](https://huggingface.co/second-state/Nous-Hermes-2-Mixtral-8x7B-DPO-GGUF/blob/main/Nous-Hermes-2-Mixtral-8x7B-DPO-Q4_K_S.gguf) | Q4_K_S | 4 | 26.7 GB| small, greater quality loss |
81
+ | [Nous-Hermes-2-Mixtral-8x7B-DPO-Q5_0.gguf](https://huggingface.co/second-state/Nous-Hermes-2-Mixtral-8x7B-DPO-GGUF/blob/main/Nous-Hermes-2-Mixtral-8x7B-DPO-Q5_0.gguf) | Q5_0 | 5 | 32.2 GB| legacy; medium, balanced quality - prefer using Q4_K_M |
82
+ | [Nous-Hermes-2-Mixtral-8x7B-DPO-Q5_K_M.gguf](https://huggingface.co/second-state/Nous-Hermes-2-Mixtral-8x7B-DPO-GGUF/blob/main/Nous-Hermes-2-Mixtral-8x7B-DPO-Q5_K_M.gguf) | Q5_K_M | 5 | 33.2 GB| large, very low quality loss - recommended |
83
+ | [Nous-Hermes-2-Mixtral-8x7B-DPO-Q5_K_S.gguf](https://huggingface.co/second-state/Nous-Hermes-2-Mixtral-8x7B-DPO-GGUF/blob/main/Nous-Hermes-2-Mixtral-8x7B-DPO-Q5_K_S.gguf) | Q5_K_S | 5 | 32.2 GB| large, low quality loss - recommended |
84
+ | [Nous-Hermes-2-Mixtral-8x7B-DPO-Q6_K.gguf](https://huggingface.co/second-state/Nous-Hermes-2-Mixtral-8x7B-DPO-GGUF/blob/main/Nous-Hermes-2-Mixtral-8x7B-DPO-Q6_K.gguf) | Q6_K | 6 | 38.4 GB| very large, extremely low quality loss |
85
+ | [Nous-Hermes-2-Mixtral-8x7B-DPO-Q8_0.gguf](https://huggingface.co/second-state/Nous-Hermes-2-Mixtral-8x7B-DPO-GGUF/blob/main/Nous-Hermes-2-Mixtral-8x7B-DPO-Q8_0.gguf) | Q8_0 | 8 | 49.6 GB| very large, extremely low quality loss - not recommended |