priorsBuilder

Mistral · released 2026-04-29

Mistral Medium 3.5

Measured on 58 benchmarks.

General frontier

17.4

#119 of 136 models · best at high

includes estimates: 24% of the mix is measured for this model

Your mix’s benchmarks

2 of 8 measured

reasoning

Humanity's Last Exam

587 results
  1. #258high of 58713.8%

Leader: Opus 5.5 · 61.4%

knowledge

AA-Omniscience

526 results
  1. #298high of 526-36.8

Leader: Opus 5.5 · 46.4

All 58 benchmarks

show 56 more

agentic

AA-AnalystAgent

44 results
  1. #37high of 4412.5%

Leader: Gemini 3.7 Flash · 60.0%

agentic

AA-Briefcase v1.1

224 results
  1. #172high of 224523

Leader: Sonnet 5.5 · 1823

agentic

Agent Arena

50 results
  1. #50high of 50-12.6%

Leader: Opus 5.5 · 15.5%

agentic

AutomationBench-AA

219 results
  1. #155high of 2196.3%

Leader: Gemini 4 · 77.5%

agentic

EnterpriseOps-Gym-AA

50 results
  1. #37high of 5033.7%

Leader: Fable 5 · 51.1%

agentic

Harvey's Legal Agent Benchmark

79 results
  1. #62high of 790.4%

Leader: Muse Spark 1.2 · 25.4%

agentic

Legal Research Bench

78 results
  1. #71high of 789.1%

Leader: Muse Spark 1.3 · 55.3%

agentic

Tax Agent Bench

69 results
  1. #66high of 694.5%

Leader: Fable 5.1 · 49.2%

agentic

τ²-Bench Telecom

393 results
  1. #37high of 39394.2%

Leader: GLM 5.2 · 99.1%

agentic

τ³-Banking

206 results
  1. #120high of 20615.1%

Leader: Qwen 3.8 Max · 51.3%

coding

Arena WebDev

122 results
  1. #110high of 1221263

Leader: Opus 5.5 · 1813

coding

Code Migration

77 results
  1. #70high of 775.1%

Leader: Sonnet 5.5 · 69.8%

coding

SciCode

226 results
  1. #168high of 22640.2%

Leader: Opus 5.5 · 66.9%

coding

SWE-bench Verified (Vals)

85 results
  1. #69high of 8566.4%

Leader: Opus 5 · 97.0%

coding

Terminal-Bench 2.0 (Vals)

63 results
  1. #44high of 6330.3%

Leader: GPT-5.5 · 73.2%

coding

Terminal-Bench 2.1 (Vals)

74 results
  1. #68high of 7439.0%

Leader: Opus 5.5 · 87.6%

coding

Terminal-Bench Hard-AA

387 results
  1. #95high of 38733.3%

Leader: GPT-5.6 Sol · 65.9%

coding

Terminal-Bench v2.1-AA

238 results
  1. #120high of 23850.6%

Leader: Fable 5.1 · 91.4%

coding

Terminal-Bench v4-AA

216 results
  1. #175high of 2160.0%

Leader: Sonnet 5.5 · 63.6%

coding

Vibe Code Bench v1.1

105 results
  1. #99high of 1052.9%

Leader: Sonnet 5.5 · 92.4%

coding

WeirdML v2

127 results
  1. #75high of 12743.7%

Leader: GPT-6 Astra Pro · 93.6%

composite

AA Intelligence Index v4.3.2

610 results
  1. #293high of 61014.2

Leader: Opus 5.5 · 57.6

composite

Arena Text

220 results
  1. #96high of 2201426

Leader: Gemini 4 · 1525

composite

Arena Text — Multi-Turn

220 results
  1. #94high of 2201432

Leader: Gemini 4 · 1553

composite

Vals Multimodal Index v1.2

33 results
  1. #33high of 3334.8%

Leader: Fable 5 · 74.2%

cost

AA Intelligence Index Cost per Task

177 results
  1. #88high of 177$0.50

Leader: GPT-6 Luna · $0.00

cost

Cost to Run AA Intelligence Index

177 results
  1. #111high of 177$1,264

Leader: GPT-6 Luna · $10.63

economics

CorpFin (Vals)

117 results
  1. #79high of 11758.8%

Leader: Opus 5 · 73.2%

economics

Excel Modeling Benchmark

74 results
  1. #72high of 7411.2%

Leader: Fable 5.1 · 76.7%

economics

Finance Agent v2

78 results
  1. #68high of 7832.1%

Leader: Gemini 4 · 65.4%

economics

GDP.pdf

210 results
  1. #177high of 2102.8%

Leader: Muse Wonder Full (Max) · 32.6%

economics

GDPval-AA v2.1

288 results
  1. #189high of 288760

Leader: Opus 5.5 · 1866

economics

MortgageTax (Vals)

84 results
  1. #83high of 8428.9%

Leader: Opus 5 · 72.1%

economics

TaxEval (Vals)

125 results
  1. #94high of 12568.0%

Leader: Muse Spark 1.2 · 80.4%

instruction

IFBench

399 results
  1. #77high of 39968.8%

Leader: Grok 4.3 · 83.3%

knowledge

AA-Omniscience Accuracy

526 results
  1. #229high of 52624.7%

Leader: Fable 5.1 · 67.2%

knowledge

AA-Omniscience Non-hallucination

526 results
  1. #294high of 52618.4%

Leader: MiniCPM5-1B (Non-reasoning) · 99.1%

knowledge

CaseLaw (Vals)

53 results
  1. #52high of 5344.2%

Leader: Grok 4.3 · 79.3%

knowledge

GPQA Diamond

539 results
  1. #250high of 53974.8%

Leader: GPT-6 Astra · 96.3%

knowledge

GPQA Diamond (Vals)

122 results
  1. #118high of 12234.8%

Leader: Gemini 3.1 Pro · 95.5%

knowledge

MedCode

94 results
  1. #76high of 9433.8%

Leader: Opus 5 · 63.6%

knowledge

MedScribe

96 results
  1. #87high of 9667.7%

Leader: Opus 5.5 · 91.4%

knowledge

MMLU Pro (Vals)

122 results
  1. #101high of 12275.3%

Leader: Fable 5.1 · 92.4%

knowledge

MMMU-Pro

265 results
  1. #174high of 26564.9%

Leader: Opus 5.5 · 87.7%

long-context

AA-LCR v1.1

523 results
  1. #224high of 52369.3%

Leader: Kimi K3 · 88.7%

long-context

MLCR-AA

94 results
  1. #86high of 941.7%

Leader: Sonnet 5.5 · 75.0%

math

ProofBench v1.1

52 results
  1. #48high of 529.0%

Leader: Sonnet 5.5 · 100.0%

other

Arena Vision

116 results
  1. #80high of 1161200

Leader: Fable 5 · 1308

reasoning

CritPt

529 results
  1. #333high of 5290.0%

Leader: Muse Wonder Full (Max) · 32.9%

reasoning

SAGE

80 results
  1. #61high of 8037.6%

Leader: Opus 4.7 · 56.1%

reliability

Tool-call error rate

171 results
  1. #35high of 1710.3%

Leader: GPT-5.4 Pro · 0.0%

reliability

Uptime (7d)

181 results
  1. #66high of 18199.8%

Leader: GPT-5 Pro · 100.0%

speed

Latency (Time to First Token)

351 results
  1. #168high of 3512.2s

Leader: Gemini 2.5 Flash-Lite · 0.3s

speed

Output Speed (tokens/s)

351 results
  1. #72high of 351167.0

Leader: Celeris-1 · 1522.5

speed

Provider latency

177 results
  1. #8high of 1770.4s

Leader: Qwen 2.5 7B · 0.2s

speed

Provider throughput

176 results
  1. #17high of 176115.0

Leader: Muse Spark 1.1 · 204.3