Mistral Small 3.2

Name: Mistral Small 3.2
Price: 0.106 USD
Author: Mistral AI

Mistral AI · Mistral Small · released 2025-06-20 · data checked 2026-11-01 Needs verification

Mistral Small 3.2 is a 24B-parameter dense model from Mistral AI (Europe), released 2025-06-20 under the Apache-2.0 license. It scores 38.6 on the Small LM Index, #31 of 34 medium models. It needs about 16.9 GB of memory at 4-bit with an 8K-token context and costs $0.11 per million tokens through the cheapest API provider.

Small LM Index

38.6

#31 of 34 scored medium models

Likely 29.0–48.9 · from 2 of 6 core benchmarks

Parameters

24B

Dense

Price per 1M tokens

$0.11

$0.075 in / $0.20 out · blended 3:1

Context window

128K

tokens

Benchmarks

Each bar shows this model's score. The tick marks the best medium model on that benchmark. Sources: 2 reported by Mistral AI. How we label sources.

MMLU-Pro Lab-reported Knowledge69.1

Class best: Qwen3.6 27B 86.2

GPQA Diamond Lab-reported Knowledge46.1

Class best: Nanbeige4.3 33B 89.5

Not reported: estimated

What the index model expects from this model's other scores. Not a measurement.

Humanity's Last Exam≈ 4

AIME≈ 50

LiveCodeBench≈ 43

SWE-bench Verified≈ 19

Specifications

Lab: Mistral AI (FR · Europe)
Architecture: Dense transformer
Total parameters: 24B
Active parameters: All (dense)
Size class: Medium
Context window: 128K tokens
Input modalities: text, image
Reasoning: Instruct (no thinking)
License: Apache-2.0
Openness: Open weights
Memory to run: ~16.9 GB at 4-bit, 8K context (more sizes)
Hugging Face: 118.9K downloads (30 days) · 643 likes
API providers: 4 on OpenRouter
API price: $0.075 in / $0.20 out per 1M (OpenRouter, 2026-11-03)

Weights on Hugging Face ↗Try on OpenRouter ↗

Something wrong here? Report it.

Run it locally

Uses unsloth/Mistral-Small-3.2-24B-Instruct-2506-GGUF (Q4_K_M). Swap the tag for another size: BF16, IQ4_NL, IQ4_XS, Q2_K, Q2_K_L, Q3_K_M, Q3_K_S, Q4_0….

Ollama

ollama run hf.co/unsloth/Mistral-Small-3.2-24B-Instruct-2506-GGUF:Q4_K_M

llama.cpp

llama-server -hf unsloth/Mistral-Small-3.2-24B-Instruct-2506-GGUF:Q4_K_M

Then open http://localhost:8080 for a chat window.

MLX (Mac)

pip install -U mlx-lm
mlx_lm.chat --model mlx-community/Mistral-Small-3.2-24B-Instruct-2506-8bit

Text only; for images use mlx-vlm.

vLLM (GPU server)

vllm serve mistralai/Mistral-Small-3.2-24B-Instruct-2506

Full-precision weights, for a server GPU.

Open in LM Studio

Memory needed: weights, context cache and runtime buffers.

Quantization	4K context	32K context	128K context
16-bit F16	50 GB	54 GB	70 GB
8-bit Q8_0	27 GB	32 GB	47 GB
4-bit Q4_K_M	16 GB	21 GB	37 GB

Check it against your GPU, Mac or phone →

Mistral Small 3.2

Benchmarks

Not reported: estimated

Specifications

Compare with

Ranked in

Sources

Run it locally