geekradar
Int
Explore GeekRadar

More to explore

Good gear All topics About us Say hello
GeekRadar / The comparison desk
COMPARE.
AI models
mistral

Mistral Small 3

AI model Released

Context window
32,768 tokens
Maximum output
16,384 tokens
Open weights
Yes

Capability & preference

Human preference · Arena

Overall text preferences from head-to-head votes. Not factual accuracy or a task-success rate.

Mistral Small 3 1233.5 mistral-small-24b-instruct-2501 · 95% interval 1227.7–1239.4 · 14,681 votes · 2026-10-15

Model performance

Radar task score: four tests, equal weight. An experimental summary of these tasks, not a universal intelligence rating.

Mistral Small 3

Radar task score

Unrated

1/4 core tests · not enough evidence

2 benchmark results for this configuration

Scientific reasoning

GPQA Diamond · 25%

Expert-written science questions; strict multiple-choice scoring. Chance accuracy is 25%.

Mistral Small 3 47.3*% Approx. 95% interval: 41.8–52.8%

Novel problem solving

Mystery Game Puzzles · 25%

Best-next-move puzzles in an undisclosed game variant. A narrow reasoning test, not general intelligence.

Mistral Small 3 Not available

Advanced mathematics

FrontierMath Tiers 1-3 · 25%

Private v2 mathematics problems with Python tools. Earlier versions and Tier 4 are excluded.

Mistral Small 3 Not available

Factual answers

SimpleQA Verified · 25%

Correct factual answers under a guessing prompt, graded by a model. Not a hallucination or abstention score.

Mistral Small 3 Not available

Competition mathematics

OTIS Mock AIME 2024-2025 · Separate result

Competition problems with model-extracted answers checked against an exact key. Earlier grading procedures are excluded.

Mistral Small 3 6.7*% Approx. 95% interval: 1.4–11.9%

Bars show reported task scores. Whiskers are approximate 95% intervals where standard errors are available. Different reasoning budgets are not equal-cost tests. Missing results are never zero.

Recorded results

Only the same benchmark, version, workload, scenario and unit share a row. Missing results are not zero. System results include the full submitted hardware and software configuration.

Arena text preference / chinesehuman preference · 2026-10-15 · rating

Mistral Small 3

1,203.66 rating

Confidence interval: 1,186.01 – 1,221.31

1054 votes

Category
chinese
Evaluated configuration
mistral-small-24b-instruct-2501
Evaluation type
human preference
Arena text preference / codinghuman preference · 2026-10-15 · rating

Mistral Small 3

1,245.55 rating

Confidence interval: 1,233.09 – 1,258.01

2083 votes

Category
coding
Evaluated configuration
mistral-small-24b-instruct-2501
Evaluation type
human preference
Arena text preference / creative_writinghuman preference · 2026-10-15 · rating

Mistral Small 3

1,194.54 rating

Confidence interval: 1,182.44 – 1,206.64

2415 votes

Category
creative_writing
Evaluated configuration
mistral-small-24b-instruct-2501
Evaluation type
human preference
Arena text preference / englishhuman preference · 2026-10-15 · rating

Mistral Small 3

1,256.09 rating

Confidence interval: 1,249.05 – 1,263.12

8740 votes

Category
english
Evaluated configuration
mistral-small-24b-instruct-2501
Evaluation type
human preference
Arena text preference / exclude_tieshuman preference · 2026-10-15 · rating

Mistral Small 3

1,129.02 rating

Confidence interval: 1,119.79 – 1,138.24

9628 votes

Category
exclude_ties
Evaluated configuration
mistral-small-24b-instruct-2501
Evaluation type
human preference
Arena text preference / experthuman preference · 2026-10-15 · rating

Mistral Small 3

1,201.61 rating

Confidence interval: 1,181.09 – 1,222.13

754 votes

Category
expert
Evaluated configuration
mistral-small-24b-instruct-2501
Evaluation type
human preference
Arena text preference / frenchhuman preference · 2026-10-15 · rating

Mistral Small 3

1,203.30 rating

Confidence interval: 1,153.96 – 1,252.65

142 votes

Category
french
Evaluated configuration
mistral-small-24b-instruct-2501
Evaluation type
human preference
Arena text preference / germanhuman preference · 2026-10-15 · rating

Mistral Small 3

1,211.35 rating

Confidence interval: 1,177.91 – 1,244.79

301 votes

Category
german
Evaluated configuration
mistral-small-24b-instruct-2501
Evaluation type
human preference
Arena text preference / hard_promptshuman preference · 2026-10-15 · rating

Mistral Small 3

1,232.97 rating

Confidence interval: 1,223.40 – 1,242.53

3569 votes

Category
hard_prompts
Evaluated configuration
mistral-small-24b-instruct-2501
Evaluation type
human preference
Arena text preference / hard_prompts_englishhuman preference · 2026-10-15 · rating

Mistral Small 3

1,242.97 rating

Confidence interval: 1,231.23 – 1,254.71

2355 votes

Category
hard_prompts_english
Evaluated configuration
mistral-small-24b-instruct-2501
Evaluation type
human preference
Arena text preference / industry_business_and_management_and_financial_operationshuman preference · 2026-10-15 · rating

Mistral Small 3

1,205.36 rating

Confidence interval: 1,189.76 – 1,220.96

1443 votes

Category
industry_business_and_management_and_financial_operations
Evaluated configuration
mistral-small-24b-instruct-2501
Evaluation type
human preference
Arena text preference / industry_entertainment_and_sports_and_mediahuman preference · 2026-10-15 · rating

Mistral Small 3

1,183.85 rating

Confidence interval: 1,172.44 – 1,195.27

2680 votes

Category
industry_entertainment_and_sports_and_media
Evaluated configuration
mistral-small-24b-instruct-2501
Evaluation type
human preference
Arena text preference / industry_legal_and_governmenthuman preference · 2026-10-15 · rating

Mistral Small 3

1,218.70 rating

Confidence interval: 1,198.04 – 1,239.36

869 votes

Category
industry_legal_and_government
Evaluated configuration
mistral-small-24b-instruct-2501
Evaluation type
human preference
Arena text preference / industry_life_and_physical_and_social_sciencehuman preference · 2026-10-15 · rating

Mistral Small 3

1,234.83 rating

Confidence interval: 1,223.09 – 1,246.58

2579 votes

Category
industry_life_and_physical_and_social_science
Evaluated configuration
mistral-small-24b-instruct-2501
Evaluation type
human preference
Arena text preference / industry_mathematicalhuman preference · 2026-10-15 · rating

Mistral Small 3

1,243.17 rating

Confidence interval: 1,228.76 – 1,257.58

1447 votes

Category
industry_mathematical
Evaluated configuration
mistral-small-24b-instruct-2501
Evaluation type
human preference
Arena text preference / industry_medicine_and_healthcarehuman preference · 2026-10-15 · rating

Mistral Small 3

1,215.45 rating

Confidence interval: 1,191.72 – 1,239.17

654 votes

Category
industry_medicine_and_healthcare
Evaluated configuration
mistral-small-24b-instruct-2501
Evaluation type
human preference
Arena text preference / industry_software_and_it_serviceshuman preference · 2026-10-15 · rating

Mistral Small 3

1,248.11 rating

Confidence interval: 1,238.20 – 1,258.02

3478 votes

Category
industry_software_and_it_services
Evaluated configuration
mistral-small-24b-instruct-2501
Evaluation type
human preference
Arena text preference / industry_writing_and_literature_and_languagehuman preference · 2026-10-15 · rating

Mistral Small 3

1,213.93 rating

Confidence interval: 1,204.50 – 1,223.37

3942 votes

Category
industry_writing_and_literature_and_language
Evaluated configuration
mistral-small-24b-instruct-2501
Evaluation type
human preference
Arena text preference / instruction_followinghuman preference · 2026-10-15 · rating

Mistral Small 3

1,213.81 rating

Confidence interval: 1,205.67 – 1,221.95

5485 votes

Category
instruction_following
Evaluated configuration
mistral-small-24b-instruct-2501
Evaluation type
human preference
Arena text preference / japanesehuman preference · 2026-10-15 · rating

Mistral Small 3

1,110.86 rating

Confidence interval: 1,074.71 – 1,147.01

269 votes

Category
japanese
Evaluated configuration
mistral-small-24b-instruct-2501
Evaluation type
human preference
Arena text preference / koreanhuman preference · 2026-10-15 · rating

Mistral Small 3

1,188.16 rating

Confidence interval: 1,148.57 – 1,227.76

231 votes

Category
korean
Evaluated configuration
mistral-small-24b-instruct-2501
Evaluation type
human preference
Arena text preference / longer_queryhuman preference · 2026-10-15 · rating

Mistral Small 3

1,245.64 rating

Confidence interval: 1,232.17 – 1,259.10

1767 votes

Category
longer_query
Evaluated configuration
mistral-small-24b-instruct-2501
Evaluation type
human preference
Arena text preference / mathhuman preference · 2026-10-15 · rating

Mistral Small 3

1,240.19 rating

Confidence interval: 1,226.83 – 1,253.54

1683 votes

Category
math
Evaluated configuration
mistral-small-24b-instruct-2501
Evaluation type
human preference
Arena text preference / multi_turnhuman preference · 2026-10-15 · rating

Mistral Small 3

1,216.61 rating

Confidence interval: 1,203.71 – 1,229.50

2151 votes

Category
multi_turn
Evaluated configuration
mistral-small-24b-instruct-2501
Evaluation type
human preference
Arena text preference / non_englishhuman preference · 2026-10-15 · rating

Mistral Small 3

1,198.14 rating

Confidence interval: 1,190.23 – 1,206.04

5941 votes

Category
non_english
Evaluated configuration
mistral-small-24b-instruct-2501
Evaluation type
human preference
Arena text preference / overallhuman preference · 2026-10-15 · rating

Mistral Small 3

1,233.53 rating

Confidence interval: 1,227.66 – 1,239.41

14681 votes

Category
overall
Evaluated configuration
mistral-small-24b-instruct-2501
Evaluation type
human preference
Arena text preference / russianhuman preference · 2026-10-15 · rating

Mistral Small 3

1,215.98 rating

Confidence interval: 1,201.33 – 1,230.63

1418 votes

Category
russian
Evaluated configuration
mistral-small-24b-instruct-2501
Evaluation type
human preference

Run it locally

Inference · memory estimate

Mistral Small 3

Not enough verified architecture information for a reliable VRAM estimate.

API cost

USD / month

Standard text-token estimate. Excludes taxes, caching, batch discounts, reasoning and multimodal surcharges. Provider limits and regional availability may differ.

Mistral Small 3

Mistral Small 3 / Pricing tiers 1
Provider / scopeRateCurrencyAmount / unit
DeepInframistralai/Mistral-Small-24B-Instruct-2501Standard InputUSD0.05 / 1M tokens
OutputUSD0.08 / 1M tokens

Capabilities

Context window
32,768 tokens
Maximum output
16,384 tokens
Reasoning
No
Tool calling
No
Open weights
Yes
Knowledge cutoff
2023-10-31
Model license
Apache 2.0

Input modalities

  • text

Output modalities

  • text

* Provisional data