---
base_model: moxin-org/moxin-chat-7b
license: apache-2.0
model_creator: moxin-org
model_name: moxin-chat-7b
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 -->

# moxin-chat-7b-GGUF

## Original Model

[moxin-org/moxin-chat-7b](https://huggingface.co/moxin-org/moxin-chat-7b)

## Run with LlamaEdge

- LlamaEdge version: coming soon

<!-- - LlamaEdge version: [v0.14.3](https://github.com/LlamaEdge/LlamaEdge/releases/tag/0.14.3) -->

- Prompt template

  - Prompt type: `moxin-chat`

  - Prompt string

    ```text
    <s> [INST] {system_message}

    {user_message_1} [/INST] {assistant_message_1}</s> [INST] {user_message_2} [/INST]
    ```
- Reverse prompt: `[INST]`

- Context size: `32000`

- Run as LlamaEdge service

  ```bash
  wasmedge --dir .:. --nn-preload default:GGML:AUTO:moxin-chat-7b-Q5_K_M.gguf \
    llama-api-server.wasm \
    --model-name moxin-chat-7b \
    --prompt-template moxin-chat \
    --reverse-prompt "[INST]" \
    --ctx-size 32000
  ```

- Run as LlamaEdge command app

  ```bash
  wasmedge --dir .:. --nn-preload default:GGML:AUTO:moxin-chat-7b-Q5_K_M.gguf \
    llama-chat.wasm \
    --prompt-template moxin-chat \
    --reverse-prompt "[INST]" \
    --ctx-size 32000
  ```

## Quantized GGUF Models

| Name | Quant method | Bits | Size | Use case |
| ---- | ---- | ---- | ---- | ----- |
| [moxin-chat-7b-Q2_K.gguf](https://huggingface.co/second-state/moxin-chat-7b-GGUF/blob/main/moxin-chat-7b-Q2_K.gguf)     | Q2_K   | 2 | 3.04 GB| smallest, significant quality loss - not recommended for most purposes |
| [moxin-chat-7b-Q3_K_L.gguf](https://huggingface.co/second-state/moxin-chat-7b-GGUF/blob/main/moxin-chat-7b-Q3_K_L.gguf) | Q3_K_L | 3 | 4.28 GB| small, substantial quality loss |
| [moxin-chat-7b-Q3_K_M.gguf](https://huggingface.co/second-state/moxin-chat-7b-GGUF/blob/main/moxin-chat-7b-Q3_K_M.gguf) | Q3_K_M | 3 | 3.94 GB| very small, high quality loss |
| [moxin-chat-7b-Q3_K_S.gguf](https://huggingface.co/second-state/moxin-chat-7b-GGUF/blob/main/moxin-chat-7b-Q3_K_S.gguf) | Q3_K_S | 3 | 3.54 GB| very small, high quality loss |
| [moxin-chat-7b-Q4_0.gguf](https://huggingface.co/second-state/moxin-chat-7b-GGUF/blob/main/moxin-chat-7b-Q4_0.gguf)     | Q4_0   | 4 | 4.60 GB| legacy; small, very high quality loss - prefer using Q3_K_M |
| [moxin-chat-7b-Q4_K_M.gguf](https://huggingface.co/second-state/moxin-chat-7b-GGUF/blob/main/moxin-chat-7b-Q4_K_M.gguf) | Q4_K_M | 4 | 4.89 GB| medium, balanced quality - recommended |
| [moxin-chat-7b-Q4_K_S.gguf](https://huggingface.co/second-state/moxin-chat-7b-GGUF/blob/main/moxin-chat-7b-Q4_K_S.gguf) | Q4_K_S | 4 | 4.63 GB| small, greater quality loss |
| [moxin-chat-7b-Q5_0.gguf](https://huggingface.co/second-state/moxin-chat-7b-GGUF/blob/main/moxin-chat-7b-Q5_0.gguf)     | Q5_0   | 5 | 5.60 GB| legacy; medium, balanced quality - prefer using Q4_K_M |
| [moxin-chat-7b-Q5_K_M.gguf](https://huggingface.co/second-state/moxin-chat-7b-GGUF/blob/main/moxin-chat-7b-Q5_K_M.gguf) | Q5_K_M | 5 | 5.75 GB| large, very low quality loss - recommended |
| [moxin-chat-7b-Q5_K_S.gguf](https://huggingface.co/second-state/moxin-chat-7b-GGUF/blob/main/moxin-chat-7b-Q5_K_S.gguf) | Q5_K_S | 5 | 5.60 GB| large, low quality loss - recommended |
| [moxin-chat-7b-Q6_K.gguf](https://huggingface.co/second-state/moxin-chat-7b-GGUF/blob/main/moxin-chat-7b-Q6_K.gguf)     | Q6_K   | 6 | 6.66 GB| very large, extremely low quality loss |
| [moxin-chat-7b-Q8_0.gguf](https://huggingface.co/second-state/moxin-chat-7b-GGUF/blob/main/moxin-chat-7b-Q8_0.gguf)     | Q8_0   | 8 | 8.62 GB| very large, extremely low quality loss - not recommended |
| [moxin-chat-7b-f16.gguf](https://huggingface.co/second-state/moxin-chat-7b-GGUF/blob/main/moxin-chat-7b-f16.gguf)       | f16    | 16 | 16.2 GB| |

*Quantized with llama.cpp b4273*