Hugging Face's logo

second-state
/
Qwen2-0.5B-GGUF

Text Generation
GGUF
English
qwen2
Model card Files Community
apepkuss79 commited on
Commit
73b642e
•
1 Parent(s): 61128c9

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +82 -82
README.md CHANGED
@@ -1,83 +1,83 @@
1
- ---
2
- base_model: Qwen/Qwen2-0.5B
3
- license: apache-2.0
4
- model_creator: Qwen
5
- model_name: Qwen2-0.5B
6
- quantized_by: Second State Inc.
7
- language:
8
- - en
9
- pipeline_tag: text-generation
10
- ---
11
-
12
- <!-- header start -->
13
- <!-- 200823 -->
14
- <div style="width: auto; margin-left: auto; margin-right: auto">
15
- <img src="https://github.com/LlamaEdge/LlamaEdge/raw/dev/assets/logo.svg" style="width: 100%; min-width: 400px; display: block; margin: auto;">
16
- </div>
17
- <hr style="margin-top: 1.0em; margin-bottom: 1.0em;">
18
- <!-- header end -->
19
-
20
- # Qwen2-0.5B-GGUF
21
-
22
- ## Original Model
23
-
24
- [Qwen/Qwen2-0.5B](https://huggingface.co/Qwen/Qwen2-0.5B)
25
-
26
- <!-- ## Run with LlamaEdge
27
-
28
- - LlamaEdge version: [v0.11.2](https://github.com/LlamaEdge/LlamaEdge/releases/tag/0.11.2)
29
-
30
- - Prompt template
31
-
32
- - Prompt type: `chatml`
33
-
34
- - Prompt string
35
-
36
- ```text
37
- <|im_start|>system
38
- {system_message}<|im_end|>
39
- <|im_start|>user
40
- {prompt}<|im_end|>
41
- <|im_start|>assistant
42
- ```
43
-
44
- - Context size: `32000`
45
-
46
- - Run as LlamaEdge service
47
-
48
- ```bash
49
- wasmedge --dir .:. --nn-preload default:GGML:AUTO:Qwen2-0.5B-Q5_K_M.gguf \
50
- llama-api-server.wasm \
51
- --model-name Qwen2-0.5B \
52
- --prompt-template chatml \
53
- --ctx-size 32000
54
- ```
55
-
56
- - Run as LlamaEdge command app
57
-
58
- ```bash
59
- wasmedge --dir .:. --nn-preload default:GGML:AUTO:Qwen2-0.5B-Q5_K_M.gguf \
60
- llama-chat.wasm \
61
- --prompt-template chatml \
62
- --ctx-size 32000
63
- ``` -->
64
-
65
- ## Quantized GGUF Models
66
-
67
- | Name | Quant method | Bits | Size | Use case |
68
- | ---- | ---- | ---- | ---- | ----- |
69
- | [Qwen2-0.5B-Q2_K.gguf](https://huggingface.co/second-state/Qwen2-0.5B-GGUF/blob/main/Qwen2-0.5B-Q2_K.gguf) | Q2_K | 2 | 339 MB| smallest, significant quality loss - not recommended for most purposes |
70
- | [Qwen2-0.5B-Q3_K_L.gguf](https://huggingface.co/second-state/Qwen2-0.5B-GGUF/blob/main/Qwen2-0.5B-Q3_K_L.gguf) | Q3_K_L | 3 | 369 MB| small, substantial quality loss |
71
- | [Qwen2-0.5B-Q3_K_M.gguf](https://huggingface.co/second-state/Qwen2-0.5B-GGUF/blob/main/Qwen2-0.5B-Q3_K_M.gguf) | Q3_K_M | 3 | 355 MB| very small, high quality loss |
72
- | [Qwen2-0.5B-Q3_K_S.gguf](https://huggingface.co/second-state/Qwen2-0.5B-GGUF/blob/main/Qwen2-0.5B-Q3_K_S.gguf) | Q3_K_S | 3 | 338 MB| very small, high quality loss |
73
- | [Qwen2-0.5B-Q4_0.gguf](https://huggingface.co/second-state/Qwen2-0.5B-GGUF/blob/main/Qwen2-0.5B-Q4_0.gguf) | Q4_0 | 4 | 352 MB| legacy; small, very high quality loss - prefer using Q3_K_M |
74
- | [Qwen2-0.5B-Q4_K_M.gguf](https://huggingface.co/second-state/Qwen2-0.5B-GGUF/blob/main/Qwen2-0.5B-Q4_K_M.gguf) | Q4_K_M | 4 | 398 MB| medium, balanced quality - recommended |
75
- | [Qwen2-0.5B-Q4_K_S.gguf](https://huggingface.co/second-state/Qwen2-0.5B-GGUF/blob/main/Qwen2-0.5B-Q4_K_S.gguf) | Q4_K_S | 4 | 385 MB| small, greater quality loss |
76
- | [Qwen2-0.5B-Q5_0.gguf](https://huggingface.co/second-state/Qwen2-0.5B-GGUF/blob/main/Qwen2-0.5B-Q5_0.gguf) | Q5_0 | 5 | 397 MB| legacy; medium, balanced quality - prefer using Q4_K_M |
77
- | [Qwen2-0.5B-Q5_K_M.gguf](https://huggingface.co/second-state/Qwen2-0.5B-GGUF/blob/main/Qwen2-0.5B-Q5_K_M.gguf) | Q5_K_M | 5 | 420 MB| large, very low quality loss - recommended |
78
- | [Qwen2-0.5B-Q5_K_S.gguf](https://huggingface.co/second-state/Qwen2-0.5B-GGUF/blob/main/Qwen2-0.5B-Q5_K_S.gguf) | Q5_K_S | 5 | 413 MB| large, low quality loss - recommended |
79
- | [Qwen2-0.5B-Q6_K.gguf](https://huggingface.co/second-state/Qwen2-0.5B-GGUF/blob/main/Qwen2-0.5B-Q6_K.gguf) | Q6_K | 6 | 506 MB| very large, extremely low quality loss |
80
- | [Qwen2-0.5B-Q8_0.gguf](https://huggingface.co/second-state/Qwen2-0.5B-GGUF/blob/main/Qwen2-0.5B-Q8_0.gguf) | Q8_0 | 8 | 531 MB| very large, extremely low quality loss - not recommended |
81
- | [Qwen2-0.5B-f16.gguf](https://huggingface.co/second-state/Qwen2-0.5B-GGUF/blob/main/Qwen2-0.5B-f16.gguf) | f16 | 16 | 994 MB| |
82
-
83
  *Quantized with llama.cpp b3613*
 
1
+ ---
2
+ base_model: Qwen/Qwen2-0.5B
3
+ license: apache-2.0
4
+ model_creator: Qwen
5
+ model_name: Qwen2-0.5B
6
+ quantized_by: Second State Inc.
7
+ language:
8
+ - en
9
+ pipeline_tag: text-generation
10
+ ---
11
+
12
+ <!-- header start -->
13
+ <!-- 200823 -->
14
+ <div style="width: auto; margin-left: auto; margin-right: auto">
15
+ <img src="https://github.com/LlamaEdge/LlamaEdge/raw/dev/assets/logo.svg" style="width: 100%; min-width: 400px; display: block; margin: auto;">
16
+ </div>
17
+ <hr style="margin-top: 1.0em; margin-bottom: 1.0em;">
18
+ <!-- header end -->
19
+
20
+ # Qwen2-0.5B-GGUF
21
+
22
+ ## Original Model
23
+
24
+ [Qwen/Qwen2-0.5B](https://huggingface.co/Qwen/Qwen2-0.5B)
25
+
26
+ ## Run with LlamaEdge
27
+
28
+ - LlamaEdge version: [v0.11.2](https://github.com/LlamaEdge/LlamaEdge/releases/tag/0.11.2)
29
+
30
+ - Prompt template
31
+
32
+ - Prompt type: `chatml`
33
+
34
+ - Prompt string
35
+
36
+ ```text
37
+ <|im_start|>system
38
+ {system_message}<|im_end|>
39
+ <|im_start|>user
40
+ {prompt}<|im_end|>
41
+ <|im_start|>assistant
42
+ ```
43
+
44
+ - Context size: `32000`
45
+
46
+ - Run as LlamaEdge service
47
+
48
+ ```bash
49
+ wasmedge --dir .:. --nn-preload default:GGML:AUTO:Qwen2-0.5B-Q5_K_M.gguf \
50
+ llama-api-server.wasm \
51
+ --model-name Qwen2-0.5B \
52
+ --prompt-template chatml \
53
+ --ctx-size 32000
54
+ ```
55
+
56
+ - Run as LlamaEdge command app
57
+
58
+ ```bash
59
+ wasmedge --dir .:. --nn-preload default:GGML:AUTO:Qwen2-0.5B-Q5_K_M.gguf \
60
+ llama-chat.wasm \
61
+ --prompt-template chatml \
62
+ --ctx-size 32000
63
+ ```
64
+
65
+ ## Quantized GGUF Models
66
+
67
+ | Name | Quant method | Bits | Size | Use case |
68
+ | ---- | ---- | ---- | ---- | ----- |
69
+ | [Qwen2-0.5B-Q2_K.gguf](https://huggingface.co/second-state/Qwen2-0.5B-GGUF/blob/main/Qwen2-0.5B-Q2_K.gguf) | Q2_K | 2 | 339 MB| smallest, significant quality loss - not recommended for most purposes |
70
+ | [Qwen2-0.5B-Q3_K_L.gguf](https://huggingface.co/second-state/Qwen2-0.5B-GGUF/blob/main/Qwen2-0.5B-Q3_K_L.gguf) | Q3_K_L | 3 | 369 MB| small, substantial quality loss |
71
+ | [Qwen2-0.5B-Q3_K_M.gguf](https://huggingface.co/second-state/Qwen2-0.5B-GGUF/blob/main/Qwen2-0.5B-Q3_K_M.gguf) | Q3_K_M | 3 | 355 MB| very small, high quality loss |
72
+ | [Qwen2-0.5B-Q3_K_S.gguf](https://huggingface.co/second-state/Qwen2-0.5B-GGUF/blob/main/Qwen2-0.5B-Q3_K_S.gguf) | Q3_K_S | 3 | 338 MB| very small, high quality loss |
73
+ | [Qwen2-0.5B-Q4_0.gguf](https://huggingface.co/second-state/Qwen2-0.5B-GGUF/blob/main/Qwen2-0.5B-Q4_0.gguf) | Q4_0 | 4 | 352 MB| legacy; small, very high quality loss - prefer using Q3_K_M |
74
+ | [Qwen2-0.5B-Q4_K_M.gguf](https://huggingface.co/second-state/Qwen2-0.5B-GGUF/blob/main/Qwen2-0.5B-Q4_K_M.gguf) | Q4_K_M | 4 | 398 MB| medium, balanced quality - recommended |
75
+ | [Qwen2-0.5B-Q4_K_S.gguf](https://huggingface.co/second-state/Qwen2-0.5B-GGUF/blob/main/Qwen2-0.5B-Q4_K_S.gguf) | Q4_K_S | 4 | 385 MB| small, greater quality loss |
76
+ | [Qwen2-0.5B-Q5_0.gguf](https://huggingface.co/second-state/Qwen2-0.5B-GGUF/blob/main/Qwen2-0.5B-Q5_0.gguf) | Q5_0 | 5 | 397 MB| legacy; medium, balanced quality - prefer using Q4_K_M |
77
+ | [Qwen2-0.5B-Q5_K_M.gguf](https://huggingface.co/second-state/Qwen2-0.5B-GGUF/blob/main/Qwen2-0.5B-Q5_K_M.gguf) | Q5_K_M | 5 | 420 MB| large, very low quality loss - recommended |
78
+ | [Qwen2-0.5B-Q5_K_S.gguf](https://huggingface.co/second-state/Qwen2-0.5B-GGUF/blob/main/Qwen2-0.5B-Q5_K_S.gguf) | Q5_K_S | 5 | 413 MB| large, low quality loss - recommended |
79
+ | [Qwen2-0.5B-Q6_K.gguf](https://huggingface.co/second-state/Qwen2-0.5B-GGUF/blob/main/Qwen2-0.5B-Q6_K.gguf) | Q6_K | 6 | 506 MB| very large, extremely low quality loss |
80
+ | [Qwen2-0.5B-Q8_0.gguf](https://huggingface.co/second-state/Qwen2-0.5B-GGUF/blob/main/Qwen2-0.5B-Q8_0.gguf) | Q8_0 | 8 | 531 MB| very large, extremely low quality loss - not recommended |
81
+ | [Qwen2-0.5B-f16.gguf](https://huggingface.co/second-state/Qwen2-0.5B-GGUF/blob/main/Qwen2-0.5B-f16.gguf) | f16 | 16 | 994 MB| |
82
+
83
  *Quantized with llama.cpp b3613*