Hugging Face's logo

second-state
/
Llama-3.3-Swallow-70B-Instruct-v0.4-GGUF

Text Generation
Transformers
GGUF
English
Japanese
llama
conversational
Model card Files
xet
Community
apepkuss79 commited on
Commit
6e77dbd
·
verified ·
1 Parent(s): 8ebf49d

Upload README.md with huggingface_hub

Browse files
Files changed (1) hide show
  1. README.md +97 -1
README.md CHANGED
@@ -1,3 +1,99 @@
1
  ---
2
- license: apache-2.0
 
 
 
 
 
 
 
 
 
 
 
 
3
  ---
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
  ---
2
+ license:
3
+ - llama3.3
4
+ - gemma
5
+ model_name: Llama-3.3-Swallow-70B-Instruct-v0.4
6
+ base_model: tokyotech-llm/Llama-3.3-Swallow-70B-Instruct-v0.4
7
+ model_creator: tokyotech-llm
8
+ quantized_by: Second State Inc.
9
+ library_name: transformers
10
+ pipeline_tag: text-generation
11
+ model_type: llama
12
+ language:
13
+ - en
14
+ - ja
15
  ---
16
+
17
+ <!-- header start -->
18
+ <!-- 200823 -->
19
+ <div style="width: auto; margin-left: auto; margin-right: auto">
20
+ <img src="https://github.com/LlamaEdge/LlamaEdge/raw/dev/assets/logo.svg" style="width: 100%; min-width: 400px; display: block; margin: auto;">
21
+ </div>
22
+ <hr style="margin-top: 1.0em; margin-bottom: 1.0em;">
23
+ <!-- header end -->
24
+
25
+ # Llama-3.3-Swallow-70B-Instruct-v0.4-GGUF
26
+
27
+ ## Original Model
28
+
29
+ [tokyotech-llm/Llama-3.3-Swallow-70B-Instruct-v0.4](https://huggingface.co/tokyotech-llm/Llama-3.3-Swallow-70B-Instruct-v0.4)
30
+
31
+ ## Run with LlamaEdge
32
+
33
+ - LlamaEdge version: [v0.16.13](https://github.com/LlamaEdge/LlamaEdge/releases/tag/0.16.13) and above
34
+
35
+ - Prompt template
36
+
37
+ - Prompt type: `llama-3-chat`
38
+
39
+ - Prompt string
40
+
41
+ ```text
42
+ <|begin_of_text|><|start_header_id|>system<|end_header_id|>
43
+
44
+ {{ system_prompt }}<|eot_id|><|start_header_id|>user<|end_header_id|>
45
+
46
+ {{ user_message_1 }}<|eot_id|><|start_header_id|>assistant<|end_header_id|>
47
+
48
+ {{ model_answer_1 }}<|eot_id|><|start_header_id|>user<|end_header_id|>
49
+
50
+ {{ user_message_2 }}<|eot_id|><|start_header_id|>assistant<|end_header_id|>
51
+ ```
52
+
53
+ - Context size: `128000`
54
+
55
+ - Run as LlamaEdge service
56
+
57
+ ```bash
58
+ wasmedge --dir .:. --nn-preload default:GGML:AUTO:Llama-3.3-Swallow-70B-Instruct-v0.4-Q5_K_M.gguf \
59
+ llama-api-server.wasm \
60
+ --prompt-template llama-3-chat \
61
+ --ctx-size 128000 \
62
+ --model-name Llama-3.3-Swallow-70B
63
+ ```
64
+
65
+ - Run as LlamaEdge command app
66
+
67
+ ```bash
68
+ wasmedge --dir .:. --nn-preload default:GGML:AUTO:Llama-3.3-Swallow-70B-Instruct-v0.4-Q5_K_M.gguf \
69
+ llama-chat.wasm \
70
+ --prompt-template llama-3-chat \
71
+ --ctx-size 128000
72
+ ```
73
+
74
+ ## Quantized GGUF Models
75
+
76
+ | Name | Quant method | Bits | Size | Use case |
77
+ | ---- | ---- | ---- | ---- | ----- |
78
+ | [Llama-3.3-Swallow-70B-Instruct-v0.4-Q2_K.gguf](https://huggingface.co/second-state/Llama-3.3-Swallow-70B-Instruct-v0.4-GGUF/blob/main/Llama-3.3-Swallow-70B-Instruct-v0.4-Q2_K.gguf) | Q2_K | 2 | 26.4 GB| smallest, significant quality loss - not recommended for most purposes |
79
+ | [Llama-3.3-Swallow-70B-Instruct-v0.4-Q3_K_L.gguf](https://huggingface.co/second-state/Llama-3.3-Swallow-70B-Instruct-v0.4-GGUF/blob/main/Llama-3.3-Swallow-70B-Instruct-v0.4-Q3_K_L.gguf) | Q3_K_L | 3 | 37.1 GB| small, substantial quality loss |
80
+ | [Llama-3.3-Swallow-70B-Instruct-v0.4-Q3_K_M.gguf](https://huggingface.co/second-state/Llama-3.3-Swallow-70B-Instruct-v0.4-GGUF/blob/main/Llama-3.3-Swallow-70B-Instruct-v0.4-Q3_K_M.gguf) | Q3_K_M | 3 | 34.3 GB| very small, high quality loss |
81
+ | [Llama-3.3-Swallow-70B-Instruct-v0.4-Q3_K_S.gguf](https://huggingface.co/second-state/Llama-3.3-Swallow-70B-Instruct-v0.4-GGUF/blob/main/Llama-3.3-Swallow-70B-Instruct-v0.4-Q3_K_S.gguf) | Q3_K_S | 3 | 30.9 GB| very small, high quality loss |
82
+ | [Llama-3.3-Swallow-70B-Instruct-v0.4-Q4_0.gguf](https://huggingface.co/second-state/Llama-3.3-Swallow-70B-Instruct-v0.4-GGUF/blob/main/Llama-3.3-Swallow-70B-Instruct-v0.4-Q4_0.gguf) | Q4_0 | 4 | 40.0 GB| legacy; small, very high quality loss - prefer using Q3_K_M |
83
+ | [Llama-3.3-Swallow-70B-Instruct-v0.4-Q4_K_M.gguf](https://huggingface.co/second-state/Llama-3.3-Swallow-70B-Instruct-v0.4-GGUF/blob/main/Llama-3.3-Swallow-70B-Instruct-v0.4-Q4_K_M.gguf) | Q4_K_M | 4 | 42.5 GB| medium, balanced quality - recommended |
84
+ | [Llama-3.3-Swallow-70B-Instruct-v0.4-Q4_K_S.gguf](https://huggingface.co/second-state/Llama-3.3-Swallow-70B-Instruct-v0.4-GGUF/blob/main/Llama-3.3-Swallow-70B-Instruct-v0.4-Q4_K_S.gguf) | Q4_K_S | 4 | 40.3 GB| small, greater quality loss |
85
+ | [Llama-3.3-Swallow-70B-Instruct-v0.4-Q5_0.gguf](https://huggingface.co/second-state/Llama-3.3-Swallow-70B-Instruct-v0.4-GGUF/blob/main/Llama-3.3-Swallow-70B-Instruct-v0.4-Q5_0.gguf) | Q5_0 | 5 | 48.7 GB| legacy; medium, balanced quality - prefer using Q4_K_M |
86
+ | [Llama-3.3-Swallow-70B-Instruct-v0.4-Q5_K_M.gguf](https://huggingface.co/second-state/Llama-3.3-Swallow-70B-Instruct-v0.4-GGUF/blob/main/Llama-3.3-Swallow-70B-Instruct-v0.4-Q5_K_M.gguf) | Q5_K_M | 5 | 49.9 GB| large, very low quality loss - recommended |
87
+ | [Llama-3.3-Swallow-70B-Instruct-v0.4-Q5_K_S.gguf](https://huggingface.co/second-state/Llama-3.3-Swallow-70B-Instruct-v0.4-GGUF/blob/main/Llama-3.3-Swallow-70B-Instruct-v0.4-Q5_K_S.gguf) | Q5_K_S | 5 | 48.7 GB| large, low quality loss - recommended |
88
+ | [Llama-3.3-Swallow-70B-Instruct-v0.4-Q6_K-00001-of-00002.gguf](https://huggingface.co/second-state/Llama-3.3-Swallow-70B-Instruct-v0.4-GGUF/blob/main/Llama-3.3-Swallow-70B-Instruct-v0.4-Q6_K-00001-of-00002.gguf) | Q6_K | 6 | 29.9 GB| very large, extremely low quality loss |
89
+ | [Llama-3.3-Swallow-70B-Instruct-v0.4-Q6_K-00002-of-00002.gguf](https://huggingface.co/second-state/Llama-3.3-Swallow-70B-Instruct-v0.4-GGUF/blob/main/Llama-3.3-Swallow-70B-Instruct-v0.4-Q6_K-00002-of-00002.gguf) | Q6_K | 6 | 28.0 GB| very large, extremely low quality loss |
90
+ | [Llama-3.3-Swallow-70B-Instruct-v0.4-Q8_0-00001-of-00003.gguf](https://huggingface.co/second-state/Llama-3.3-Swallow-70B-Instruct-v0.4-GGUF/blob/main/Llama-3.3-Swallow-70B-Instruct-v0.4-Q8_0-00001-of-00003.gguf) | Q8_0 | 8 | 29.9 GB| very large, extremely low quality loss - not recommended |
91
+ | [Llama-3.3-Swallow-70B-Instruct-v0.4-Q8_0-00002-of-00003.gguf](https://huggingface.co/second-state/Llama-3.3-Swallow-70B-Instruct-v0.4-GGUF/blob/main/Llama-3.3-Swallow-70B-Instruct-v0.4-Q8_0-00002-of-00003.gguf) | Q8_0 | 8 | 29.8 GB| very large, extremely low quality loss - not recommended |
92
+ | [Llama-3.3-Swallow-70B-Instruct-v0.4-Q8_0-00003-of-00003.gguf](https://huggingface.co/second-state/Llama-3.3-Swallow-70B-Instruct-v0.4-GGUF/blob/main/Llama-3.3-Swallow-70B-Instruct-v0.4-Q8_0-00003-of-00003.gguf) | Q8_0 | 8 | 15.3 GB| very large, extremely low quality loss - not recommended |
93
+ | [Llama-3.3-Swallow-70B-Instruct-v0.4-f16-00001-of-00005.gguf](https://huggingface.co/second-state/Llama-3.3-Swallow-70B-Instruct-v0.4-GGUF/blob/main/Llama-3.3-Swallow-70B-Instruct-v0.4-f16-00001-of-00005.gguf) | f16 | 16 | 30.0 GB| |
94
+ | [Llama-3.3-Swallow-70B-Instruct-v0.4-f16-00002-of-00005.gguf](https://huggingface.co/second-state/Llama-3.3-Swallow-70B-Instruct-v0.4-GGUF/blob/main/Llama-3.3-Swallow-70B-Instruct-v0.4-f16-00002-of-00005.gguf) | f16 | 16 | 29.6 GB| |
95
+ | [Llama-3.3-Swallow-70B-Instruct-v0.4-f16-00003-of-00005.gguf](https://huggingface.co/second-state/Llama-3.3-Swallow-70B-Instruct-v0.4-GGUF/blob/main/Llama-3.3-Swallow-70B-Instruct-v0.4-f16-00003-of-00005.gguf) | f16 | 16 | 29.9 GB| |
96
+ | [Llama-3.3-Swallow-70B-Instruct-v0.4-f16-00004-of-00005.gguf](https://huggingface.co/second-state/Llama-3.3-Swallow-70B-Instruct-v0.4-GGUF/blob/main/Llama-3.3-Swallow-70B-Instruct-v0.4-f16-00004-of-00005.gguf) | f16 | 16 | 29.6 GB| |
97
+ | [Llama-3.3-Swallow-70B-Instruct-v0.4-f16-00005-of-00005.gguf](https://huggingface.co/second-state/Llama-3.3-Swallow-70B-Instruct-v0.4-GGUF/blob/main/Llama-3.3-Swallow-70B-Instruct-v0.4-f16-00005-of-00005.gguf) | f16 | 16 | 22.2 GB| |
98
+
99
+ *Quantized with llama.cpp b4944.*