Modelpublic

second-state/Mistral-Small-24B-Instruct-2501-GGUF

sourceHugging Faceapache-2.0updated 2y agoView on Hugging Face

0likes666downloads

Model Card

<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;">

Mistral-Small-24B-Instruct-2501-GGUF

Original Model

mistralai/Mistral-Small-24B-Instruct-2501

Run with LlamaEdge

—LlamaEdge version: v0.16.5

—Prompt template

—Chat

—Prompt type: mistral-small-chat

—Prompt string

text

      <s>[SYSTEM_PROMPT]<system prompt>[/SYSTEM_PROMPT][INST]<user message>[/INST]<assistant response></s>[INST]<user message>[/INST]

—Chat + Tool Use

—Prompt type: mistral-small-tool

—Prompt string

text

      <s>[INST] {user_message_1}[/INST][TOOL_CALLS] [{tool_call_1},{tool_call_2}]</s>[TOOL_RESULTS] {tool_result_1}[/TOOL_RESULTS] {assistant_message_1}</s>[AVAILABLE_TOOLS] [{tool_1},{tool_2}][/AVAILABLE_TOOLS][INST] {system_message}<0x0A><0x0A>{user_message_2}[/INST]

—Context size: 32000

—Run as LlamaEdge service

—Chat

bash

    wasmedge --dir .:. --nn-preload default:GGML:AUTO:Mistral-Small-24B-Instruct-2501-Q5_K_M.gguf \
        llama-api-server.wasm \
        --prompt-template mistral-small-chat \
        --ctx-size 32000 \
        --model-name Mistral-Small-24B-Instruct-2501

—Chat + Tool Use

bash

    wasmedge --dir .:. --nn-preload default:GGML:AUTO:Mistral-Small-24B-Instruct-2501-Q5_K_M.gguf \
        llama-api-server.wasm \
        --prompt-template mistral-small-tool \
        --ctx-size 32000 \
        --model-name Mistral-Small-24B-Instruct-2501

Example:

image/png

—Run as LlamaEdge command app

—Chat

bash

    wasmedge --dir .:. --nn-preload default:GGML:AUTO:Mistral-Small-24B-Instruct-2501-Q5_K_M.gguf \
      llama-chat.wasm \
      --prompt-template mistral-small-chat \
      --ctx-size 32000

Quantized GGUF Models

Name	Quant method	Bits	Size	Use case
Mistral-Small-24B-Instruct-2501-Q2_K.gguf	Q2_K	2	8.89 GB	smallest, significant quality loss - not recommended for most purposes
Mistral-Small-24B-Instruct-2501-Q3_K_L.gguf	Q3KL	3	12.4 GB	small, substantial quality loss
Mistral-Small-24B-Instruct-2501-Q3_K_M.gguf	Q3KM	3	11.5 GB	very small, high quality loss
Mistral-Small-24B-Instruct-2501-Q3_K_S.gguf	Q3KS	3	10.4 GB	very small, high quality loss
Mistral-Small-24B-Instruct-2501-Q4_0.gguf	Q4_0	4	13.4 GB	legacy; small, very high quality loss - prefer using Q3KM
Mistral-Small-24B-Instruct-2501-Q4_K_M.gguf	Q4KM	4	14.3 GB	medium, balanced quality - recommended
Mistral-Small-24B-Instruct-2501-Q4_K_S.gguf	Q4KS	4	13.5 GB	small, greater quality loss
Mistral-Small-24B-Instruct-2501-Q5_0.gguf	Q5_0	5	16.3 GB	legacy; medium, balanced quality - prefer using Q4KM
Mistral-Small-24B-Instruct-2501-Q5_K_M.gguf	Q5KM	5	16.8 GB	large, very low quality loss - recommended
Mistral-Small-24B-Instruct-2501-Q5_K_S.gguf	Q5KS	5	16.3 GB	large, low quality loss - recommended
Mistral-Small-24B-Instruct-2501-Q6_K.gguf	Q6_K	6	19.3 GB	very large, extremely low quality loss
Mistral-Small-24B-Instruct-2501-Q8_0.gguf	Q8_0	8	25.1 GB	very large, extremely low quality loss - not recommended
Mistral-Small-24B-Instruct-2501-f16.gguf	f16	16	47.2 GB

Quantized with llama.cpp b4595.