Model reference · open weights

Mistral-Small-3.2-2506

Available as managed deployment LLMs mistralai Vision + text 1 variants 40k dl/mo

Mistral-Small-3.2-2506 is an open-weight language model from mistralai. AxForge deploys and operates it for you on dedicated EU-owned hardware — with the licence handled where one is required.

Available as managed deployment — configured and operated for you on dedicated EU hardware, quoted per deployment.

What it is

Released byMistral AI
Published undermistralai
TypeLanguage models
TaskVision + text
Runs withvllm
Based onmistralai/Mistral-Small-3.2-24B-Instruct-2506
Released2025-06-20
Popularity40k downloads / month
LicenceOpen weights

About

What Mistral-Small-3.2-2506 is

[!NOTE] Includes our GGUF chat template fixes! Tool calling works as well! If you are using llama.cpp, use --jinja to enable the system prompt.

Run in llama.cpp:

./llama.cpp/llama-cli -hf unsloth/Mistral-Small-3.2-24B-Instruct-2506-GGUF:UD-Q4_K_XL --jinja --temp 0.15 --top-k -1 --top-p 1.00 -ngl 99

Run in Ollama:

ollama run hf.co/unsloth/Mistral-Small-3.2-24B-Instruct-2506-GGUF:UD-Q4_K_XL
  • Temperature of: 0.15
  • Set top_p to: 1.00
  • Max tokens (context length): 128K
Read the full model card

Mistral-Small-3.2-24B-Instruct-2506

Mistral-Small-3.2-24B-Instruct-2506 is a minor update of Mistral-Small-3.1-24B-Instruct-2503.

Small-3.2 improves in the following categories:

  • Instruction following: Small-3.2 is better at following precise instructions
  • Repetition errors: Small-3.2 produces less infinite generations or repetitive answers
  • Function calling: Small-3.2's function calling template is more robust (see here and examples)

In all other categories Small-3.2 should match or slightly improve compared to Mistral-Small-3.1-24B-Instruct-2503.

Key Features

Benchmark Results

We compare Mistral-Small-3.2-24B to Mistral-Small-3.1-24B-Instruct-2503. For more comparison against other models of similar size, please check Mistral-Small-3.1's Benchmarks'

Text

Instruction Following / Chat / Tone
ModelWildbench v2Arena Hard v2IF (Internal; accuracy)
Small 3.1 24B Instruct55.6%19.56%82.75%
Small 3.2 24B Instruct65.33%43.1%84.78%
Infinite Generations

Small 3.2 reduces infitine generations by 2x on challenging, long and repetitive prompts.

ModelInfinite Generations (Internal; Lower is better)
Small 3.1 24B Instruct2.11%
Small 3.2 24B Instruct1.29%
STEM
ModelMMLUMMLU Pro (5-shot CoT)MATHGPQA Main (5-shot CoT)GPQA Diamond (5-shot CoT )MBPP Plus - Pass@5HumanEval Plus - Pass@5SimpleQA (TotalAcc)
Small 3.1 24B Instruct80.62%66.76%69.30%44.42%45.96%74.63%88.99%10.43%
Small 3.2 24B Instruct80.50%69.06%69.42%44.22%46.13%78.33%92.90%12.10%

Vision

ModelMMMUMathvistaChartQADocVQAAI2D
Small 3.1 24B Instruct64.00%68.91%86.24%94.08%93.72%
Small 3.2 24B Instruct62.50%67.09%87.4%94.86%92.91%

Usage

The model can be used with the following frameworks;

Note 1: We recommend using a relatively low temperature, such as temperature=0.15.

Note 2: Make sure to add a system prompt to the model to best tailer it for your needs. If you want to use the model as a general assistant, we recommend to use the one provided in the SYSTEM_PROMPT.txt file.

vLLM (recommended)

We recommend using this model with vLLM.

Installation

Make sure to install vLLM >= 0.9.1:

pip install vllm --upgrade

Doing so should automatically install mistral_common >= 1.6.2.

To check:

python -c "import mistral_common; print(mistral_common.__version__)"

You can also make use of a ready-to-go docker image or on the docker hub.

Serve

We recommand that you use Mistral-Small-3.2-24B-Instruct-2506 in a server/client setting.

  1. Spin up a server:
vllm serve mistralai/Mistral-Small-3.2-24B-Instruct-2506 --tokenizer_mode mistral --config_format mistral --load_format mistral --tool-call-parser mistral --enable-auto-tool-choice --limit_mm_per_prompt 'image=10' --tensor-parallel-size 2

Note: Running Mistral-Small-3.2-24B-Instruct-2506 on GPU requires ~55 GB of GPU RAM in bf16 or fp16.

  1. To ping the client you can use a simple Python snippet. See the following examples.
Vision reasoning

Take leverage of the vision capabilities of Mistral-Small-3.2-24B-Instruct-2506 to take the best choice given a scenario, go catch them all !

from datetime import datetime, timedelta

from openai import OpenAI
from huggingface_hub import hf_hub_download

From the published model card. Full card on the HuggingFace links in the sidebar.

How it works

How language models work

Your prompttext / messagesTransformerattention over tokensNext-token loopgenerate + streamResponsetext · tool callsA language model reads your tokens and predicts the next one, again and again, streaming the reply back.

Using it via the API

Call it like any OpenAI endpoint

Once AxForge deploys mistral-small-3-2-2506 for you, it answers on the OpenAI-compatible API — the same base URL and keys as every other model. (mistral-small-3-2-2506 below is illustrative; you get the exact model name on deployment.)

$ curl -sS https://api.axforge.ai/v1/chat/completions \
  -H "Authorization: Bearer $AXFORGE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"mistral-small-3-2-2506","messages":[{"role":"user","content":"Hello"}]}'

Create an account — your API key is available in the console. 3M free tokens every 30 days with every new account.

© 2026 AxForge · EU-hosted AI infrastructure Pricing Docs Trust Privacy Terms