Mistral Small 3.2 24B
Mistral · mistralai/Mistral-Small-3.2-24B-Instruct-2506
Q4 file 14.3 GB (Q4_K_M) · runs well from 24 GB · Weights license: Apache 2.0
Quality
- LMArena · text · rating
- 1,356 ± 5Confidence interval 1,351–1,361 · 17,454 votes · rank 212 of the release · Arena lists Mistral Small 3.2 (2506), earlier under Mistral's API name mistral-small-2506; Mistral's own model card for Mistral Small 3.2 (version 25.06) gives that API name and these open weightsLMArena 2026-10-22 · CC BY 4.0
No result in LiveBench 2026-06-25 Not in LiveBench 2026-06-25
Reported by the publisher
- Wildbench v2
- 65.33%Instruction Following / Chat / Tone
- Arena Hard v2
- 43.1%Instruction Following / Chat / Tone
- IF (Internal; accuracy)
- 84.78%Instruction Following / Chat / Tone
- Infinite Generations (Internal; Lower is better)
- 1.29%Infinite Generations
- MMLU
- 80.50%STEM
- MMLU Pro (5-shot CoT)
- 69.06%STEM · 5-shot
- MATH
- 69.42%STEM
- GPQA Main (5-shot CoT)
- 44.22%STEM · 5-shot
Every reported result
Values as printed on the card, with the publisher's own settings (card section: Benchmark Results › Text › Instruction Following / Chat / Tone; Benchmark Results › Text › Infinite Generations); column “Small 3.2 24B Instruct”. Not comparable across publishers and not used in any ranking. Model card
Run it
Run it
File Q4_K_M · 14.3 GB · unsloth/Mistral-Small-3.2-24B-Instruct-2506-GGUF · What is Q4?
Measured 19–101 tok/s · Q4_K_M · 6 machines · Local LLM Advisor · all rows
ollama run hf.co/unsloth/Mistral-Small-3.2-24B-Instruct-2506-GGUF:Q4_K_M- Install Ollama
- Run the command above
- Ask:
Explain what a context window is, in two sentences.
Runs on
Fast memory holds it 48
The planning estimate fits the memory the model can use quickly: graphics card VRAM, or unified memory within the macOS GPU share.
Show all 48
Within memory, outside the fast part 45
Within the machine's memory, outside its fast part: on a graphics card the layers that do not fit run on the CPU from PC RAM (the listed RAM, or 32 GB assumed); on a Mac, above the macOS default GPU share. Not a speed figure.
Show all 45
One memory tier short 9
The same machines with the next memory size up would hold it in fast memory.
Show all 9
Mistral Small 3.2 24B is a text model from Mistral with 24B parameters. It was released on June 20, 2025 and reads up to 131,072 tokens of context. Its Q4 weights plus working room come to about 19 GB before the machine's reserve; the smallest reference machines whose fast memory holds that are Mac mini M4 · 32 GB unified (~27 GB with its reserve) and RTX 4090 · 24 GB VRAM (~21 GB with its reserve). We collected 29 downloadable GGUF and MLX files from 3 publisher sources; the most downloaded GGUF source is unsloth's. It has no result in LiveBench 2026-06-25. Its LMArena text rating is 1,356 ± 5 (release 2026-10-22), a separate human-preference scale.
Hardware profiles grouped by the Q4 planning estimate on each machine (dedicated GPUs reserve 2 GB; unified and system memory 25%, 2 to 8 GB): the fast memory holds it, it is within memory but outside the fast part (layers offloaded to PC RAM, or above the macOS GPU share), or the machine is one memory tier short. Each card opens its hardware profile. Not a measured fit or speed.
On the reference machines
| Where Q4 sits | ||||
|---|---|---|---|---|
| Mac mini M2 · 8 GB unifiedLarger than its memoryNo file on the site's quant ladder fits the fast memory at 8,192 tokensNo file in the previous column | ~21 GB | Larger than its memory | No file on the site's quant ladder fits the fast memory at 8,192 tokens | No file in the previous column |
| Mac mini M4 · 16 GB unifiedLarger than its memoryNo file on the site's quant ladder fits the fast memory at 8,192 tokensNo file in the previous column | ~23 GB | Larger than its memory | No file on the site's quant ladder fits the fast memory at 8,192 tokens | No file in the previous column |
| Mac mini M4 · 24 GB unifiedLarger than its memoryIQ4_XS8,192 tokens | ~25 GB | Larger than its memory | IQ4_XS | 8,192 tokens |
| Mac mini M4 · 32 GB unifiedFast memory holds itQ5_K_M16,384 tokens | ~27 GB | Fast memory holds it | Q5_K_M | 16,384 tokens |
| Mac mini M5 Pro · 64 GB unifiedFast memory holds itQ8_065,536 tokens | ~27 GB | Fast memory holds it | Q8_0 | 65,536 tokens |
| Mac Studio M5 Max · 128 GB unifiedFast memory holds itQ8_0131,072 tokens | ~27 GB | Fast memory holds it | Q8_0 | 131,072 tokens |
| Mac Studio · M5 Ultra · 36-core CPU / 80-core GPU · 512 GB unifiedFast memory holds itQ8_0131,072 tokens | ~27 GB | Fast memory holds it | Q8_0 | 131,072 tokens |
| RTX 4060 · 8 GB VRAMIn memory, outside the fast partNo file on the site's quant ladder fits the fast memory at 8,192 tokensNo file in the previous column | ~21 GB | In memory, outside the fast part | No file on the site's quant ladder fits the fast memory at 8,192 tokens | No file in the previous column |
| RTX 3060 · 12 GB VRAMIn memory, outside the fast partNo file on the site's quant ladder fits the fast memory at 8,192 tokensNo file in the previous column | ~21 GB | In memory, outside the fast part | No file on the site's quant ladder fits the fast memory at 8,192 tokens | No file in the previous column |
| RTX 5080 · 16 GB VRAMIn memory, outside the fast partNo file on the site's quant ladder fits the fast memory at 8,192 tokensNo file in the previous column | ~21 GB | In memory, outside the fast part | No file on the site's quant ladder fits the fast memory at 8,192 tokens | No file in the previous column |
| RTX 4090 · 24 GB VRAMFast memory holds itQ5_K_M8,192 tokens | ~21 GB | Fast memory holds it | Q5_K_M | 8,192 tokens |
| RTX 5090 · 32 GB VRAMFast memory holds itQ6_K32,768 tokens | ~21 GB | Fast memory holds it | Q6_K | 32,768 tokens |
| DGX H100 · 8 × 80 GBFast memory holds itQ8_0131,072 tokens | ~35 GB | Fast memory holds it | Q8_0 | 131,072 tokens |
The file column uses the site's quant ladder (Q8_0 to Q3_K_M, MLX 8 to 4-bit on a Mac): the largest collected file whose size × 1.2 plus the context cache and the machine's reserve fits the fast memory. Planning estimates, not measured runs.
Compare Mistral Small 3.2 24B using its publisher card. No overall score appears in the LiveBench 2026-06-25 snapshot.
The Q4 weights plus working room come to about 19 GB from 24.01B stated parameters, before the machine's reserve (Runs on adds it per machine). It is not a measured minimum; context, batch size, runtime and quantization change actual memory use.
Versions to download
Files you can download for this checkpoint. The section opens on the most downloaded collected GGUF source, unsloth, at Q4_K_M. On a Mac it opens on the most downloaded MLX source. Sizes are the Hub's published bytes. What is Q4?
unsloth
unsloth/Mistral-Small-3.2-24B-Instruct-2506-GGUF · Mistral-Small-3.2-24B-Instruct-2506-Q4_K_M.gguf · 14.3 GB
Run this file
ollama run hf.co/unsloth/Mistral-Small-3.2-24B-Instruct-2506-GGUF:Q4_K_Mllama-server -hf unsloth/Mistral-Small-3.2-24B-Instruct-2506-GGUF:Q4_K_Mlms get https://huggingface.co/unsloth/Mistral-Small-3.2-24B-Instruct-2506-GGUF- mlx-lm: mlx-lm runs MLX repositories; pick the MLX version on a Mac
Quality scores on this page belong to the original checkpoint; these files are not scored or ranked. File sizes read 2026-10-23 · download counts read 2026-10-17 · commands: Hugging Face Hub docs: Use Ollama with any GGUF model · llama.cpp server README (-hf <user>/<model>[:quant]) · LM Studio docs: lms get with a Hugging Face URL · mlx-lm README (mlx_lm.generate --model)
Fine-tunes, adapters and other versions
- Quantized versions
- 72
- Fine-tunes
- 66
- Adapters · merges
- 19 · 4
Browse them on the Hugging Face model tree · read 2026-10-23
Uncensored and abliterated versions
Some community uploads remove the model's safety refusals. Quality may drop, the original publisher does not support them, claims about them are the uploader's, and the original license still applies. This site never recommends or ranks them.
4 derivative repositories have “uncensored” or “abliterated” in their name (read 2026-10-23).
Measured speed
Tokens per second by machine
Output tok/s · 6 machines · fastest 5 of 7 · every row in the table · What is a token?
| RTX 5090 | 101.14 ± 0.24 | Q4_K_M | llama.cpp CUDA | unsloth/Mistral-Small-3.2-24B-Instruct-2506-GGUF/Mistral-Small-3.2-24B-Instruct-2506-Q4_K_M.gguf |
|---|---|---|---|---|
| RTX 5090 | 63.85 ± 0.10 | Q8_0 | llama.cpp CUDA | unsloth/Mistral-Small-3.2-24B-Instruct-2506-GGUF/Mistral-Small-3.2-24B-Instruct-2506-Q8_0.gguf |
| RTX 4090 24 GB | 61.56 ± 0.07 | Q4_K_M | llama.cpp CUDA | unsloth/Mistral-Small-3.2-24B-Instruct-2506-GGUF/Mistral-Small-3.2-24B-Instruct-2506-Q4_K_M.gguf |
| RTX 5080 16 GB | 58.64 ± 0.05 | Q4_K_M | llama.cpp CUDA | unsloth/Mistral-Small-3.2-24B-Instruct-2506-GGUF/Mistral-Small-3.2-24B-Instruct-2506-Q4_K_M.gguf |
| RTX 3090 24 GB | 53.15 ± 0.01 | Q4_K_M | llama.cpp CUDA | unsloth/Mistral-Small-3.2-24B-Instruct-2506-GGUF/Mistral-Small-3.2-24B-Instruct-2506-Q4_K_M.gguf |
| RTX 5070 Ti 16 GB | 52.85 ± 0.07 | Q4_K_M | llama.cpp CUDA | unsloth/Mistral-Small-3.2-24B-Instruct-2506-GGUF/Mistral-Small-3.2-24B-Instruct-2506-Q4_K_M.gguf |
| RTX 4060 Ti 16GB | 19.08 ± 0.00 | Q4_K_M | llama.cpp CUDA | unsloth/Mistral-Small-3.2-24B-Instruct-2506-GGUF/Mistral-Small-3.2-24B-Instruct-2506-Q4_K_M.gguf |
Source: Data: Local LLM Advisor (localllmadvisor.com) · CC BY 4.0 (checked 2026-10-25) · llama.cpp CUDA · llama.cpp commit 3e4dfa8f3 (b11658) · not estimated
How to run
The memory needed above is the Q4 load before a machine's reserve. Formula: parameters × 0.625 bytes, plus 20% working allowance; each machine then adds its system reserve: 25% of unified or system memory, 2 to 8 GB, or 2 GB per GPU on dedicated cards. This is neither measured VRAM nor checkpoint download size.
Q4 planning estimate only. Check checkpoint format, runtime support, KV cache, context, batch size and offload before choosing hardware.
Evidence level: Generic estimate only; no execution reproduced by this site.
- Publisher sampling defaults
- temperature 0.15generation_config.json · read 2026-10-23source
- Chat template markers
- — Chat template not read for this checkpoint
Before downloading
- Check the exact checkpoint and license linked here.
- Confirm a compatible runtime and precision or quantization for your platform.
- Allow for context, batch size and any extra media components beyond the weights.
Technical details
What the source and the catalog store for this checkpoint. Missing fields are left out, never zero.
- Checkpoint / version
- mistralai/Mistral-Small-3.2-24B-Instruct-2506
- Task and modality
- language and reasoning · text + image
- Count and basis
- 24B · Hub safetensors countmistralai/Mistral-Small-3.2-24B-Instruct-2506 Hugging Face Hub metadata: safetensors parameter count · read 2026-10-17source
- Context
- 131,072 tokens published for this modelmistralai/Mistral-Small-3.2-24B-Instruct-2506 config.json (Hugging Face) · read 2026-10-17source
- Architecture
- Densemistralai/Mistral-Small-3.2-24B-Instruct-2506 config.json (Hugging Face) · read 2026-10-17source
- Released
- Mistral platform-docs-public changelog 2025-06-20 (GitHub) · read 2026-10-17source
- Runtime and precision
- No local execution verified.
- Languages
- en, fr, de, es, pt, it, ja, ko, ru, zh, ar, fa, id, ms, ne, pl, ro, sr, sv, tr, uk, vi, hi, bnModel card metadata · read 2026-10-25source
- Commercial use
- Allowed Apache License 2.0 text (SPDX) · read 2026-10-18license text
Every stored field
- Revision checked
- 95a6d26c4bfb886c58daf9d3f7332c857cb27b43
- Architecture
- Mistral3ForConditionalGeneration
- Library
- vllm
- Declared base models
- mistralai/Mistral-Small-3.1-24B-Base-2503
- Formats
- safetensors
- Access
- Open download
- Publisher kind
- Official
- Repository created
- 2025-06-19creation is not the model's release
- Repository updated
- 2025-12-22a commit is not a release date
- Hub downloads (30 days) · likes
- 181,449 · 640this repository only · read 2026-10-25
- Catalog state
- Hub config or weights found; no inventory benchmark link
The 24.0B is the Hugging Face Hub's safetensors count for this repository, not a publisher figure. The 3.2 card states no parameter count and calls this a minor update of Mistral Small 3.1, whose card says 24 billion parameters. The Q4 estimate on this page is a generic planning figure, not a measured requirement.
Sources
Identity and published specifications come from the checkpoint page. Benchmark scores belong to the named release and do not measure speed on your computer.
- Publisher checkpoint and model card
- Weight license and use terms
- Quality benchmark source · LiveBench 2026-06-25 overall
Catalog reviewed: 2026-10-17. Benchmark release: 2026-06-25; snapshot collected: 2026-10-18.



























































