Skip to main content

Model profile

Mistral Small 4

MistralCurrentReleased Feb 20, 2026
Data verified
Overall Score
46.23Public #140 of 200
Arena Elo
1358
Eligible category ranks
2of 8
Price (1M tokens)
$0.15 in / $0.6 out
API pricing
Speed
175tok/s
Context
256K

Evidence coverage

16 of 323 tracked benchmarks are published. 0 are verified and 16 provisional. 6 of 8 categories are measured.

Updated July 23, 2026Methodology
Published / tracked
16 / 323
Verified
0
Provisional
16
Categories with evidence
6 / 8

Evidence by category

  • Agentic4 benchmarks
    Reported
  • Coding2 benchmarks
    Reported
  • Reasoning2 benchmarks
    Reported
  • Knowledge6 benchmarks
    Reported
  • Math0 benchmarks
    Not measured
  • Multilingual0 benchmarks
    Not measured
  • Multimodal1 benchmark
    Reported
  • Inst. Following1 benchmark
    Reported
Open WeightSelf-hostNon-Reasoning
Confidence:
Low
base

Mistral Small 4 ranks #140 out of 200 models on the public leaderboard with an overall score of 46.23/100. It does not yet have enough sourced coverage for BenchLM's verified leaderboard. While not a frontier model, it offers specific advantages depending on the use case.

Mistral Small 4 is a open weight model with a 256K token context window. It processes queries without explicit chain-of-thought reasoning, offering faster response times and lower token usage.

Mistral Small 4 sits inside the Mistral Small 4 family alongside Mistral Small 4 (Reasoning). This profile currently has 16 of 323 tracked benchmarks. BenchLM only exposes non-generated benchmark rows publicly, so missing categories stay blank until a sourced evaluation is available.

Its strongest category is Agentic (#88), while its weakest is Coding (#89). This performance profile makes it particularly useful for coding agents, browser research, and computer-use workflows.

Peer position

Exact provisional scores and ranks for the closest listed peers. A score can appear before a model clears the evidence threshold for a rank, so equal scores can have different rank states.

Range 45.1347.29

  1. Gemma 4 12B
    Google
    #13747.29
    Gemma 4 12B is #137 with a score of 47.29.
    Compare
  2. Qwen3.5 Plus
    Alibaba
    #13847.2
    Qwen3.5 Plus is #138 with a score of 47.2.
    Compare
  3. GPT-5 nano
    OpenAI
    #13946.36
    GPT-5 nano is #139 with a score of 46.36.
    Compare
  4. Mistral Small 4Current model
    Mistral
    #14046.23
    Mistral Small 4 is #140 with a score of 46.23.
  5. o1-pro
    OpenAI
    #14145.94
    o1-pro is #141 with a score of 45.94.
    Compare
  6. Claude 4.1 Opus
    Anthropic
    #14245.88
    Claude 4.1 Opus is #142 with a score of 45.88.
    Compare
  7. Z-1
    Z
    #14345.13
    Z-1 is #143 with a score of 45.13.
    Compare

Category percentile

More

Relative position among models eligible for each sourced category. A higher percentile means a stronger position within that category's ranked cohort; 100 is highest.

  1. Agentic26%
    Eligible cohort rank #88 of 119Category score 42.3
  2. Coding27%
    Eligible cohort rank #89 of 122Category score 46.3

Category evidence

Scores and ranks appear only where this model has published benchmark evidence. Categories without displayable source records remain not measured.

Category scores, ranks, weighting, benchmark coverage, and evidence status
CategoryScore
AgenticRank #88 of 119Percentile 26thWeight 22%4 benchmarksReported42.3
CodingRank #89 of 122Percentile 27thWeight 20%2 benchmarksReported46.3
ReasoningWeight 17%2 benchmarksReportedScore pending
KnowledgeRank Not rankedWeight 12%6 benchmarksReported62.3
MathWeight 5%0 benchmarksNot measuredNot measured
MultilingualWeight 7%0 benchmarksNot measuredNot measured
MultimodalWeight 12%1 benchmarkReportedScore pending
Inst. FollowingRank Not rankedWeight 5%1 benchmarkReported72.9

Self-host vs API cost

Estimates at 50,000 req/day · 1000 tokens/req average.

Mistral Small 4
API / mo$563
Self-host / mo$2,278
Break-even266M/day
Model the full break-even

Chatbot Arena performance

Scroll horizontally to inspect confidence intervals and vote counts.

Chatbot Arena Elo, confidence interval, and vote count by evaluation view
ViewEloConfidence intervalVotes
Text Overall1358±5.217,696
Coding1413±10.03,359
Math1339±17.51,041
Instruction Following1339±8.94,383
Creative Writing1325±12.82,167
Multi-turn1367±10.63,082
Hard Prompts1374±6.97,813
Hard Prompts (English)1398±9.14,071
Longer Query1360±10.13,401

Benchmark Details

Rows below have a displayable published verification record. Each source link and provenance note remains in the page HTML while its category is closed. Source-unverified manual rows and generated rows stay hidden.

Agentic4 benchmarks
AA Agentic IndexReported

Artificial Analysis Agentic Index

4.7%Display only
Source: Artificial Analysis model benchmarksProvenance: Display-only row synced from the current Artificial Analysis model payload. It is excluded from BenchLM weighted scoring.
τ²-bench resultsReported

τ²-Bench Tool-Agent-User Evaluation

41.2%Display only
Source: Artificial Analysis model benchmarksProvenance: Display-only row synced from the current Artificial Analysis model payload. It is excluded from BenchLM weighted scoring.
GDPval-AAReported

GDPval-AA normalized

4.4%Display only
Source: Artificial Analysis model benchmarksProvenance: Display-only row synced from the current Artificial Analysis model payload. It is excluded from BenchLM weighted scoring.
GDPval-AAReported
588Display only
Source: Artificial Analysis model benchmarksProvenance: Display-only row synced from the current Artificial Analysis model payload. It is excluded from BenchLM weighted scoring.
Coding2 benchmarks
AA Coding IndexReported

Artificial Analysis Coding Index

26.6%Display only
Source: Artificial Analysis model benchmarksProvenance: Display-only row synced from the current Artificial Analysis model payload. It is excluded from BenchLM weighted scoring.
AA-SciCodeReported

Artificial Analysis SciCode

38.0%Display only
Source: Artificial Analysis model benchmarksProvenance: Display-only row synced from the current Artificial Analysis model payload. It is excluded from BenchLM weighted scoring.
Reasoning2 benchmarks
AA-LCRReported

Artificial Analysis Long Context Reasoning

44.7%Display only
Source: Artificial Analysis model benchmarksProvenance: Display-only row synced from the current Artificial Analysis model payload. It is excluded from BenchLM weighted scoring.
CritPtReported

Critical Physics Tasks

0.3%Display only
Source: Artificial Analysis model benchmarksProvenance: Display-only row synced from the current Artificial Analysis model payload. It is excluded from BenchLM weighted scoring.
Knowledge6 benchmarks
Artificial Analysis Intelligence IndexReported
19.6%Display only
Source: Artificial Analysis model benchmarksProvenance: Display-only row synced from the current Artificial Analysis model payload. It is excluded from BenchLM weighted scoring.
AA-GPQA DiamondReported

Artificial Analysis GPQA Diamond

76.9%Display only
Source: Artificial Analysis model benchmarksProvenance: Display-only row synced from the current Artificial Analysis model payload. It is excluded from BenchLM weighted scoring.
AA-HLEReported

Artificial Analysis Humanity's Last Exam

9.5%Display only
Source: Artificial Analysis model benchmarksProvenance: Display-only row synced from the current Artificial Analysis model payload. It is excluded from BenchLM weighted scoring.
AA-Omniscience IndexReported

Artificial Analysis Omniscience Index

-29.9%Display only
Source: Artificial Analysis model benchmarksProvenance: Display-only row synced from the current Artificial Analysis model payload. It is excluded from BenchLM weighted scoring.
AA-Omniscience AccuracyReported

Artificial Analysis Omniscience Accuracy

22.1%Display only
Source: Artificial Analysis model benchmarksProvenance: Display-only row synced from the current Artificial Analysis model payload. It is excluded from BenchLM weighted scoring.
AA-Omniscience Hallucination RateReported

Artificial Analysis Omniscience Hallucination Rate

66.8%Display only
Source: Artificial Analysis model benchmarksProvenance: Display-only row synced from the current Artificial Analysis model payload. It is excluded from BenchLM weighted scoring.
Multimodal1 benchmark
AA-MMMU-ProReported

Artificial Analysis MMMU-Pro

56.8%Display only
Source: Artificial Analysis model benchmarksProvenance: Display-only row synced from the current Artificial Analysis model payload. It is excluded from BenchLM weighted scoring.
Inst. Following1 benchmark
AA-IFBenchReported

Artificial Analysis IFBench

48.2%Display only
Source: Artificial Analysis model benchmarksProvenance: Display-only row synced from the current Artificial Analysis model payload. It is excluded from BenchLM weighted scoring.

Mistral Small 4 Family

Base entry

Frequently Asked Questions

How does Mistral Small 4 perform overall in AI benchmarks?

Mistral Small 4 has 16 published benchmark scores on BenchLM, but it does not yet have enough non-generated coverage to receive a global overall rank.

Is Mistral Small 4 good for knowledge and understanding?

Mistral Small 4 has visible benchmark coverage in knowledge and understanding, but BenchLM does not currently assign it a global category rank there.

Is Mistral Small 4 good for coding and programming?

Mistral Small 4 ranks #89 out of 122 models in coding and programming benchmarks with an average score of 46.3. There are stronger options in this category.

Is Mistral Small 4 good for reasoning and logic?

Mistral Small 4 has visible benchmark coverage in reasoning and logic, but BenchLM does not currently assign it a global category rank there.

Is Mistral Small 4 good for agentic tool use and computer tasks?

Mistral Small 4 ranks #88 out of 119 models in agentic tool use and computer tasks benchmarks with an average score of 42.3. There are stronger options in this category.

Is Mistral Small 4 good for multimodal and grounded tasks?

Mistral Small 4 has visible benchmark coverage in multimodal and grounded tasks, but BenchLM does not currently assign it a global category rank there.

Is Mistral Small 4 good for instruction following?

Mistral Small 4 has visible benchmark coverage in instruction following, but BenchLM does not currently assign it a global category rank there.

Is Mistral Small 4 open source?

Yes, Mistral Small 4 is an open weight model created by Mistral, meaning it can be downloaded and run locally or fine-tuned for specific use cases.

Which sibling models are related to Mistral Small 4?

Mistral Small 4 belongs to the Mistral Small 4 family. Related variants on BenchLM include Mistral Small 4 (Reasoning).

Does Mistral Small 4 have full benchmark coverage on BenchLM?

Not yet. Mistral Small 4 currently has 16 published benchmark scores out of the 323 benchmarks BenchLM tracks. BenchLM only exposes non-generated public benchmark rows, so missing categories stay blank until a sourced evaluation is available.

What is the context window size of Mistral Small 4?

Mistral Small 4 has a published context window of 256K, which determines how much text it can process in a single interaction.

Last updated: July 23, 2026 · Runtime metrics stay blank until BenchLM has a sourced snapshot.

Choose with this week’s evidence

Join 2,000+ readers for ranking moves, new releases, pricing changes, and the evidence behind them.

Free. One email per week.