whatmodel.sh / Mistral Small 3 24B (2501)

Mistral Small 3 24B (2501)

+ contribute
Update model informationAdd a place to run itAdd a test resultPropose a benchmarkCorrect a benchmark scoreEdit data on GitHub

Mistral Small 3 24B (2501) is an open-weights language model from Mistral, released January 2025. Its published size is 23.57B parameters and a 32k-token context window. Licence: apache-2.0.

Catalogue identity notes

Comparison model from the Epoch AI internal-evaluation CSVs (Epoch release date 2025-01-25, 2025-01-30). Mistral Small 3 (version 25.01). Epoch labels 'mistral-small-2501' (Mistral API id; 'Mistral Small 3.0' 25.01 in Mistral's legacy table, retired 11/30/2025) and 'mistral-small-3-2501' both sit in model_metadata group 'Mistral Small 3' and name the same 2501 release, so the selection rule is applied across both labels. Distinct from mistral-small-3-2 (2506). 23,572,403,200 parameters.

53.2±1.0#293 of 719 LLMsfrom 18 benchmark results across 12 sources

$ how it compares

The LLMs ranked either side of Mistral Small 3 24B (2501), out of 828 with a score, then more from Mistral.

#modelscorewhereRAMreleased
348DeepSeek-V2 Chat (0628) DeepSeek54±1local~66GB2024-07
349GLM-4 9B Chat 1M Zhipu AI (THUDM)53±5local~8.1GB2024-06
350Mistral Small 3 24B (2501) Mistral53±1local~16GB2025-01
351Qwen2-VL 72B Instruct Alibaba53±1local~34GB2024-09
352Gemini 1.5 Flash Experimental (0827) Google53±1comparison only2024-08
more from Mistral
79Mistral Large 4 (public preview) Mistral71±2api $2.06/M2026-10
174Mistral Medium 3.5 (128B) Mistral65±1api $3.00/M +local~42GB2026-04
180Mistral Medium 3.1 (2508) Mistral64±2comparison only2025-08

see all LLMs · find the ones that fit your machine

$ what it's good at

coding▁▁▁▃▆██▇▆▄▃▁ #235 of 458 * · 4.6% of its score · 1 test

Beats Qwen3-VL 32B Instruct, Solar Open 100B (102B-A12B), GLM-4.7-Flash here, though they rank higher overall. Trails QwQ-32B-Preview, InternLM2.5 20B Chat, Gemma 2 27B IT here, though they rank lower overall.

testresultvs the others in its rundifficultycountswhy
WebApp1K-React pass@1 (onekq-ai)55.5% ✎▃▁▁█▁▄▅▃▄▇▁▄ beats 12/31medium · avg 44%▇▇ 4.6%sharp · at its level · noisy
chat & writing▁▁▁▃▅▄▅█▆▆▇▃ #251 of 393 * · 10% of its score · 1 test

Beats Mistral Small 4 (119B-A6.5B), Ministral 3 14B Reasoning 2512, Qwen2.5 Coder 32B Instruct here, though they rank higher overall. Trails Flan-T5 XXL, Granite 4.0 H 1B, Granite 4.0 1B here, though they rank lower overall.

testresultvs the others in its rundifficultycountswhy
LMArena (text)1274.4 ✎▁▁▁▂▃▃▃▆▆▇█▃ beats 106/379avg 1268▇▇▇▇ 10%blunt
math▁▂▃▅▆▅▄█▄▁▁▁ #265 of 398 * · 30% of its score · 2 tests

Beats Molmo 2 8B, Llama 3.3 70B Instruct, DeepSeek-V2.5 here, though they rank higher overall. Trails DeepSeek R1 Distill Qwen 1.5B, Mixtral 8x7B v0.1 (base), InternVL2-4B here, though they rank lower overall.

testresultvs the others in its rundifficultycountswhy
MATH level 544.8% ✎▂▂▅▆▄█▆▃▅▂▃▄ beats 36/92hard · avg 32%▇▇▇▇▇▇▇▇▇▇ 24%sharp · at its level
OTIS mock AIME 2024–20256.7% ✎▃▃▄▃▃▃▇▇█▃▂▅ beats 38/193very hard · avg 8.5%▇▇ 5.2%sharp · noisy
other tests14 not in a task · 56% of its score
testresultvs the others in its rundifficultycountswhy
GPQA Diamond47.3% ✎▁▁▁▁▂▄▆▄▅▇█▅ beats 57/208hard · avg 40%▇▇▇▇▇▇ 14%sharp · at its level
CPTUB average (SpeakLeash, 0-5 judge scale)74.2% ✎▁▂▂▂▃█▃▆▅▄▂▃ beats 66/95easy · avg 66%▇▇ 6.1%at its level
EQ-Bench 3 (emotional intelligence, judge Elo)1058.9 ✎▁·▂▁▄▅▅▃█▄▁▁ beats 7/62avg 1044▇▇ 6.0%typical
BRIDGE clinical text, zero-shot average (87 tasks)37.5% ✎▁▁▁▁▁▂▂▃▄▆█▃ beats 57/97hard · avg 32%▇▇ 5.1%blunt · at its level
AraGen-v3 3C3H score (Inception, Claude Sonnet 4 judge)39.8% ✎▁▁·▁▂▂▄█▄▅▄▄ beats 36/66hard · avg 31%▇▇ 5.0%at its level
HindiGen-v1 3C3H score (Inception, Claude 3.5 Sonnet judge)50.3% ✎▁▁▁▁·▂·█▆█▇▁ beats 7/31medium · avg 48%▇▇ 5.0%at its level
Polish EQ-Bench v2 (SpeakLeash)70.5% ✎▁··▁▁▂▂▆▅▆█▆ beats 53/70medium · avg 64%▇▇ 4.9%at its level
Judgemark v2 (LLM-as-judge for creative writing)63.6% ✎▃▃··▃▅▅···█▃ beats 5/10medium · avg 55%▇ 3.6%at its level
6 more tests
testresultvs the others in its rundifficultycountswhy
Open PT-LLM Leaderboard average (9 Portuguese tasks)77.2% ✎▃▁▁▁▁▁▁▁▂█▅▁ beats 239/276medium · avg 73%▇ 2.0%at its level · noisy
EuroEval HellaSwag (English), test split, 5-shot generative MC, accuracy91.0% ✎▃·▁▁▂▁▂▅█▇▄▁ beats 95/100medium · avg 68%▇ 1.6%at its level · noisy
EuroEval Life in the UK (English knowledge MC), test split, 5-shot, accuracy94.3% ✎▁▁▁▁▁▁▁▂▄█▄▃ beats 136/146easy · avg 83%▇ 0.9%noisy
Chess Puzzles (Epoch)0.0% ✎█▁▁▁▁▂▁▄▂▂▁▁ beats 5/135very hard · avg 5.9%▇ 0.6%sharp · too hard for it · noisy
Vectara hallucination leaderboard, 2025 dataset (HHEM-2.3, factual consistency rate)96.9% ✎▁▁▂▃▄▄█▆▃▂▂▁ beats 100/140very easy · avg 94%▇ 0.5%blunt · maxed out · noisy
Vectara hallucination leaderboard (HHEM-2.3, factual consistency rate)94.9% ✎▁▂▁▂▅█▂▃▄▃▁▁ beats 89/98very easy · avg 90%▇ 0.0%blunt

$ where to run it

llama.cpp · recommended

version

  1. Install llama.cpp · Homebrew and winget installs name the commands llama-cli and llama-server instead.
    macOS / Linux
    curl -LsSf https://llama.app/install.sh | sh
    Windows
    irm https://llama.app/install.ps1 | iex

    other ways to install ↗

  2. Run Mistral Small 3 24B (2501) · needs ~14.3 GB · from huggingface.co/lmstudio-community
    chat in the terminal
    llama cli -hf lmstudio-community/Mistral-Small-24B-Instruct-2501-GGUF:Q4_K_M
  3. Or start a local server and open the address it prints (http://localhost:9931)
    llama serve -hf lmstudio-community/Mistral-Small-24B-Instruct-2501-GGUF:Q4_K_M

LM Studio

version

  1. Install LM Studio · Open the app once before using the lms command.

    download it (macOS / Windows / Linux) ↗

  2. Run Mistral Small 3 24B (2501) · needs ~14.3 GB · from huggingface.co/lmstudio-community

    open in LM Studio ↗

  3. Or download it in the terminal
    lms get https://huggingface.co/lmstudio-community/Mistral-Small-24B-Instruct-2501-GGUF@Q4_K_M -y
  4. Then run lms ls to see its name, and lms chat followed by that name

Ollama

version

  1. Install Ollama

    download it (macOS / Windows) ↗

    Linux
    curl -fsSL https://ollama.com/install.sh | sh
  2. Run Mistral Small 3 24B (2501) · needs ~14.64 GB · from ollama.com
    chat in the terminal
    ollama run mistral-small:24b-instruct-2501-q4_K_M

MLX (Mac)

version

  1. Install MLX (Mac) · Apple silicon Macs only, text only.
    with uv
    uv tool install mlx-lm
    or
    pip install mlx-lm
  2. Run Mistral Small 3 24B (2501) · needs ~13.3 GB, Mac only · from huggingface.co/mlx-community
    chat in the terminal
    mlx_lm.chat --model mlx-community/Mistral-Small-24B-Instruct-2501-4bit
  3. Or start an OpenAI-compatible server at http://localhost:8080/v1
    mlx_lm.server --model mlx-community/Mistral-Small-24B-Instruct-2501-4bit

Weights and documentation: huggingface.co/mistralai ↗

$ how the score was worked out · Epoch AI — internal evaluatio… 45%, Inception Arabic/Hindi leader… 10%, LMArena leaderboard dataset 10% and 9 more

Each result implies a score on its own, after adjusting for how hard its test is and who ran it. The score is their weighted average, with a small pull toward the middle for thinly tested models. Each test's weight is its counts column above; here it's added up by who ran the tests. How it works

sourceresultsshare of the scorelicence
Epoch AI — internal evaluations4▇▇▇▇▇▇▇▇▇▇▇▇▇ 45%CC BY 4.0
Inception Arabic/Hindi leaderboards (AraGen-v3, HindiGen-v1, Arabic IFEval)2▇▇▇ 10%Apache-2.0
LMArena leaderboard dataset1▇▇▇ 10%CC BY 4.0
SpeakLeash CPTUB (Complex Polish Text Understanding Benchmark)1▇▇ 6%MIT
EQ-Bench 3 canonical leaderboard results1▇▇ 6%MIT
BRIDGE Medical Leaderboard (Harvard Medical School / Brigham and Women's YLab)1▇▇ 5%MIT
SpeakLeash Polish EQ-Bench leaderboard1▇ 5%Apache-2.0
WebApp1K models leaderboard (onekq-ai)1▇ 5%MIT
Judgemark v2 results (EQ-Bench)1▇ 4%MIT
EuroEval leaderboards (results collection), English generative leaderboard2▇ 3%MIT
Open Portuguese LLM Leaderboard (archived)1▇ 2%Apache-2.0
Vectara hallucination leaderboard (vectara/hallucination-leaderboard repository)2▇ 1%Apache-2.0
prior · pull toward the middle▇ 0%
weighted average= 53.2

$ improve this page

Update model informationAdd a place to run itAdd a test resultPropose a benchmarkCorrect a benchmark scoreEdit data on GitHub

Use a short GitHub form or edit the data through a pull request. Every number needs a source; here's which ones we can use.

data updated · sources and licences · json