Hugging Face's logo

second-state
/
internlm3-8b-instruct-GGUF

Text Generation
GGUF
internlm3
custom_code
conversational
Model card Files
xet
 Community
internlm3-8b-instruct-GGUF
File size: 4,779 Bytes
ed10435
 
c58791c
 
 
 
 
ed10435
c58791c
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
3ab54c7
 
 
 
 
 
 
 
 
 
 
 
 
c58791c
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
---

license: apache-2.0
pipeline_tag: text-generation
base_model: internlm/internlm3-8b-instruct
model_creator: InternLM
model_name: internlm3-8b-instruct
quantized_by: Second State Inc.
---


<!-- header start -->
<!-- 200823 -->
<div style="width: auto; margin-left: auto; margin-right: auto">
<img src="https://github.com/LlamaEdge/LlamaEdge/raw/dev/assets/logo.svg" style="width: 100%; min-width: 400px; display: block; margin: auto;">
</div>
<hr style="margin-top: 1.0em; margin-bottom: 1.0em;">
<!-- header end -->

# internlm3-8b-instruct-GGUF

## Original Model

[internlm/internlm3-8b-instruct](https://huggingface.co/internlm/internlm3-8b-instruct)

## Run with LlamaEdge

- LlamaEdge version: [v0.16.1](https://github.com/LlamaEdge/LlamaEdge/releases/tag/0.16.1) and above

- Prompt template

  - Prompt type: `chatml`

  - Prompt string

    ```text

    <|im_start|>system

    {system_message}<|im_end|>

    <|im_start|>user

    {prompt}<|im_end|>

    <|im_start|>assistant

    ```


- Context size: `128000`

- Run as LlamaEdge service

  - Chat

    ```bash

    wasmedge --dir .:. --nn-preload default:GGML:AUTO:internlm3-8b-instruct-Q5_K_M.gguf \

      llama-api-server.wasm \

      --prompt-template chatml \

      --ctx-size 128000 \

      --model-name internlm3-8b-instruct

    ```


  - Tool use

    ```bash

    wasmedge --dir .:. --nn-preload default:GGML:AUTO:internlm3-8b-instruct-Q5_K_M.gguf \

      llama-api-server.wasm \

      --prompt-template internlm-2-tool \

      --ctx-size 128000 \

      --model-name internlm3-8b-instruct

    ```


- Run as LlamaEdge command app

  ```bash

  wasmedge --dir .:. \

    --nn-preload default:GGML:AUTO:internlm3-8b-instruct-Q5_K_M.gguf \

    llama-chat.wasm \

    --prompt-template chatml \

    --ctx-size 32000

  ```

## Quantized GGUF Models

| Name | Quant method | Bits | Size | Use case |
| ---- | ---- | ---- | ---- | ----- |
| [internlm3-8b-instruct-Q2_K.gguf](https://huggingface.co/second-state/internlm3-8b-instruct-GGUF/blob/main/internlm3-8b-instruct-Q2_K.gguf)     | Q2_K   | 2 | 3.45 GB| smallest, significant quality loss - not recommended for most purposes |

| [internlm3-8b-instruct-Q3_K_L.gguf](https://huggingface.co/second-state/internlm3-8b-instruct-GGUF/blob/main/internlm3-8b-instruct-Q3_K_L.gguf) | Q3_K_L | 3 | 4.73 GB| small, substantial quality loss |

| [internlm3-8b-instruct-Q3_K_M.gguf](https://huggingface.co/second-state/internlm3-8b-instruct-GGUF/blob/main/internlm3-8b-instruct-Q3_K_M.gguf) | Q3_K_M | 3 | 4.39 GB| very small, high quality loss |

| [internlm3-8b-instruct-Q3_K_S.gguf](https://huggingface.co/second-state/internlm3-8b-instruct-GGUF/blob/main/internlm3-8b-instruct-Q3_K_S.gguf) | Q3_K_S | 3 | 3.99 GB| very small, high quality loss |

| [internlm3-8b-instruct-Q4_0.gguf](https://huggingface.co/second-state/internlm3-8b-instruct-GGUF/blob/main/internlm3-8b-instruct-Q4_0.gguf)     | Q4_0   | 4 | 5.09 GB| legacy; small, very high quality loss - prefer using Q3_K_M |
| [internlm3-8b-instruct-Q4_K_M.gguf](https://huggingface.co/second-state/internlm3-8b-instruct-GGUF/blob/main/internlm3-8b-instruct-Q4_K_M.gguf) | Q4_K_M | 4 | 5.36 GB| medium, balanced quality - recommended |
| [internlm3-8b-instruct-Q4_K_S.gguf](https://huggingface.co/second-state/internlm3-8b-instruct-GGUF/blob/main/internlm3-8b-instruct-Q4_K_S.gguf) | Q4_K_S | 4 | 5.12 GB| small, greater quality loss |
| [internlm3-8b-instruct-Q5_0.gguf](https://huggingface.co/second-state/internlm3-8b-instruct-GGUF/blob/main/internlm3-8b-instruct-Q5_0.gguf)     | Q5_0   | 5 | 6.13 GB| legacy; medium, balanced quality - prefer using Q4_K_M |

| [internlm3-8b-instruct-Q5_K_M.gguf](https://huggingface.co/second-state/internlm3-8b-instruct-GGUF/blob/main/internlm3-8b-instruct-Q5_K_M.gguf) | Q5_K_M | 5 | 6.26 GB| large, very low quality loss - recommended |

| [internlm3-8b-instruct-Q5_K_S.gguf](https://huggingface.co/second-state/internlm3-8b-instruct-GGUF/blob/main/internlm3-8b-instruct-Q5_K_S.gguf) | Q5_K_S | 5 | 6.13 GB| large, low quality loss - recommended |

| [internlm3-8b-instruct-Q6_K.gguf](https://huggingface.co/second-state/internlm3-8b-instruct-GGUF/blob/main/internlm3-8b-instruct-Q6_K.gguf)     | Q6_K   | 6 | 7.23 GB| very large, extremely low quality loss |
| [internlm3-8b-instruct-Q8_0.gguf](https://huggingface.co/second-state/internlm3-8b-instruct-GGUF/blob/main/internlm3-8b-instruct-Q8_0.gguf)     | Q8_0   | 8 | 9.36 GB| very large, extremely low quality loss - not recommended |

| [internlm3-8b-instruct-f16.gguf](https://huggingface.co/second-state/internlm3-8b-instruct-GGUF/blob/main/internlm3-8b-instruct-f16.gguf)       | f16    | 16 | 17.6 GB|  |



*Quantized with llama.cpp b4497*