Hugging Face's logo

second-state
/
Sky-T1-32B-Preview-GGUF

Text Generation
Transformers
GGUF
English
qwen2
Inference Endpoints
conversational
Model card Files Community
apepkuss79 commited on
Commit
32ba757
·
verified ·
1 Parent(s): db25edc

Upload README.md with huggingface_hub

Browse files
Files changed (1) hide show
  1. README.md +84 -1
README.md CHANGED
@@ -1,3 +1,86 @@
1
  ---
2
- license: apache-2.0
 
 
 
 
 
 
3
  ---
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
  ---
2
+ base_model: NovaSky-AI/Sky-T1-32B-Preview
3
+ model_creator: NovaSky-AI
4
+ model_name: Sky-T1-32B-Preview
5
+ quantized_by: Second State Inc.
6
+ library_name: transformers
7
+ language:
8
+ - en
9
  ---
10
+
11
+ <!-- header start -->
12
+ <!-- 200823 -->
13
+ <div style="width: auto; margin-left: auto; margin-right: auto">
14
+ <img src="https://github.com/LlamaEdge/LlamaEdge/raw/dev/assets/logo.svg" style="width: 100%; min-width: 400px; display: block; margin: auto;">
15
+ </div>
16
+ <hr style="margin-top: 1.0em; margin-bottom: 1.0em;">
17
+ <!-- header end -->
18
+
19
+ # Sky-T1-32B-Preview-GGUF
20
+
21
+ ## Original Model
22
+
23
+ [NovaSky-AI/Sky-T1-32B-Preview](https://huggingface.co/NovaSky-AI/Sky-T1-32B-Preview)
24
+
25
+ ## Run with LlamaEdge
26
+
27
+ - LlamaEdge version: coming soon
28
+
29
+ <!-- - LlamaEdge version: [v0.14.3](https://github.com/LlamaEdge/LlamaEdge/releases/tag/0.14.3) -->
30
+
31
+ - Prompt template
32
+
33
+ - Prompt type: `chatml`
34
+
35
+ - Prompt string
36
+
37
+ ```text
38
+ <|im_start|>system
39
+ {system_message}<|im_end|>
40
+ <|im_start|>user
41
+ {prompt}<|im_end|>
42
+ <|im_start|>assistant
43
+ ```
44
+
45
+ - Context size: `128000`
46
+
47
+ - Run as LlamaEdge service
48
+
49
+ ```bash
50
+ wasmedge --dir .:. --nn-preload default:GGML:AUTO:Sky-T1-32B-Preview-Q5_K_M.gguf \
51
+ llama-api-server.wasm \
52
+ --model-name Sky-T1-32B-Preview \
53
+ --prompt-template chatml \
54
+ --ctx-size 128000
55
+ ```
56
+
57
+ - Run as LlamaEdge command app
58
+
59
+ ```bash
60
+ wasmedge --dir .:. --nn-preload default:GGML:AUTO:Sky-T1-32B-Preview-Q5_K_M.gguf \
61
+ llama-chat.wasm \
62
+ --prompt-template chatml \
63
+ --ctx-size 128000
64
+ ```
65
+
66
+ ## Quantized GGUF Models
67
+
68
+ | Name | Quant method | Bits | Size | Use case |
69
+ | ---- | ---- | ---- | ---- | ----- |
70
+ | [Sky-T1-32B-Preview-Q2_K.gguf](https://huggingface.co/second-state/Sky-T1-32B-Preview-GGUF/blob/main/Sky-T1-32B-Preview-Q2_K.gguf) | Q2_K | 2 | 12.3 GB| smallest, significant quality loss - not recommended for most purposes |
71
+ | [Sky-T1-32B-Preview-Q3_K_L.gguf](https://huggingface.co/second-state/Sky-T1-32B-Preview-GGUF/blob/main/Sky-T1-32B-Preview-Q3_K_L.gguf) | Q3_K_L | 3 | 17.2 GB| small, substantial quality loss |
72
+ | [Sky-T1-32B-Preview-Q3_K_M.gguf](https://huggingface.co/second-state/Sky-T1-32B-Preview-GGUF/blob/main/Sky-T1-32B-Preview-Q3_K_M.gguf) | Q3_K_M | 3 | 15.9 GB| very small, high quality loss |
73
+ | [Sky-T1-32B-Preview-Q3_K_S.gguf](https://huggingface.co/second-state/Sky-T1-32B-Preview-GGUF/blob/main/Sky-T1-32B-Preview-Q3_K_S.gguf) | Q3_K_S | 3 | 14.4 GB| very small, high quality loss |
74
+ | [Sky-T1-32B-Preview-Q4_0.gguf](https://huggingface.co/second-state/Sky-T1-32B-Preview-GGUF/blob/main/Sky-T1-32B-Preview-Q4_0.gguf) | Q4_0 | 4 | 18.6 GB| legacy; small, very high quality loss - prefer using Q3_K_M |
75
+ | [Sky-T1-32B-Preview-Q4_K_M.gguf](https://huggingface.co/second-state/Sky-T1-32B-Preview-GGUF/blob/main/Sky-T1-32B-Preview-Q4_K_M.gguf) | Q4_K_M | 4 | 19.9 GB| medium, balanced quality - recommended |
76
+ | [Sky-T1-32B-Preview-Q4_K_S.gguf](https://huggingface.co/second-state/Sky-T1-32B-Preview-GGUF/blob/main/Sky-T1-32B-Preview-Q4_K_S.gguf) | Q4_K_S | 4 | 18.8 GB| small, greater quality loss |
77
+ | [Sky-T1-32B-Preview-Q5_0.gguf](https://huggingface.co/second-state/Sky-T1-32B-Preview-GGUF/blob/main/Sky-T1-32B-Preview-Q5_0.gguf) | Q5_0 | 5 | 22.6 GB| legacy; medium, balanced quality - prefer using Q4_K_M |
78
+ | [Sky-T1-32B-Preview-Q5_K_M.gguf](https://huggingface.co/second-state/Sky-T1-32B-Preview-GGUF/blob/main/Sky-T1-32B-Preview-Q5_K_M.gguf) | Q5_K_M | 5 | 23.3 GB| large, very low quality loss - recommended |
79
+ | [Sky-T1-32B-Preview-Q5_K_S.gguf](https://huggingface.co/second-state/Sky-T1-32B-Preview-GGUF/blob/main/Sky-T1-32B-Preview-Q5_K_S.gguf) | Q5_K_S | 5 | 22.6 GB| large, low quality loss - recommended |
80
+ | [Sky-T1-32B-Preview-Q6_K.gguf](https://huggingface.co/second-state/Sky-T1-32B-Preview-GGUF/blob/main/Sky-T1-32B-Preview-Q6_K.gguf) | Q6_K | 6 | 26.9 GB| very large, extremely low quality loss |
81
+ | [Sky-T1-32B-Preview-Q8_0.gguf](https://huggingface.co/second-state/Sky-T1-32B-Preview-GGUF/blob/main/Sky-T1-32B-Preview-Q8_0.gguf) | Q8_0 | 8 | 34.8 GB| very large, extremely low quality loss - not recommended |
82
+ | [Sky-T1-32B-Preview-f16-00001-of-00003.gguf](https://huggingface.co/second-state/Sky-T1-32B-Preview-GGUF/blob/main/Sky-T1-32B-Preview-f16-00001-of-00003.gguf) | f16 | 16 | 29.8 GB| |
83
+ | [Sky-T1-32B-Preview-f16-00002-of-00003.gguf](https://huggingface.co/second-state/Sky-T1-32B-Preview-GGUF/blob/main/Sky-T1-32B-Preview-f16-00002-of-00003.gguf) | f16 | 16 | 29.8 GB| |
84
+ | [Sky-T1-32B-Preview-f16-00003-of-00003.gguf](https://huggingface.co/second-state/Sky-T1-32B-Preview-GGUF/blob/main/Sky-T1-32B-Preview-f16-00003-of-00003.gguf) | f16 | 16 | 5.87 GB| |
85
+
86
+ *Quantized with llama.cpp b4462*