Skip to content
aiqbench

Mistral · model passport

Mistral Medium 2505

mistral-medium-2505
  • Cloud only
  • Proprietary licence
  • Released 2025-05-01(from model name)
11/27axes covered
Overall · LMArena
1,387#167 of 396
Best percentile
58.1Overall
Cohort standing
43.5median 48.9

What the data says

Benchmarked on 11 of 27 tracked axes. Every figure below is the model's own published result on that board.

Weakest axes

  • 8.7 pct — FrontierMath (0.3%)
  • 10.9 pct — Humanity's Last Exam (4.5%)

Uneven: 49.4 percentile points between its best and worst axis — 58.1 pct on Overall versus 8.7 pct on FrontierMath.

  • Overall58.1 pct1,387Verify →
  • BFCL v4 (function calling)Dormant board57.3 pct37.7%Verify →
  • Coding57.3 pct1,432Verify →
  • Hard prompts56.8 pct1,401Verify →
  • Instruction following55.6 pct1,366Verify →
  • LiveBench45 pct50.3%Verify →
  • GPQA Diamond42.1 pct59.5%Verify →
  • AIME (math)34 pct32.2%Verify →
  • Vision Arena33.3 pct1,155Verify →
  • Humanity's Last Exam10.9 pct4.5%Verify →
  • FrontierMath8.7 pct0.3%Verify →
  • WebDev ArenaNot benchmarked
  • Factuality ArenaNot benchmarked
  • Search ArenaNot benchmarked
  • Document ArenaNot benchmarked
  • Agent ArenaNot benchmarked
  • SimpleQA VerifiedNot benchmarked
  • ARC-AGI-2Not benchmarked
  • SimpleBenchNot benchmarked
  • SWE-bench Verified (live board)Not benchmarked
  • Terminal-Bench 4.0Not benchmarked
  • MMLU-Pro (Artificial Analysis)Not benchmarked
  • LiveCodeBench (Artificial Analysis)Not benchmarked
  • IFBench (Artificial Analysis)Not benchmarked
  • HHEM Hallucination (Vectara)Not benchmarked
  • SWE-bench VerifiedNot benchmarked
  • Terminal-BenchNot benchmarked

Percentiles are the model's standing within each board's own population. Axes marked "not benchmarked" have no published result for this model — that is an absence of data, not a low score.

  • mistral-medium-2505 stands at 43.5 (raw mean 41.7) — measured on 11 of 27 axes.

    standingtemplateVerify →

  • mistral-medium-2505 stands at 43.5 versus a cohort median of 48.9 across 148 models released within 180 days — below the peer median.

    cohorttemplateVerify →

Every claim is a computed fact from the current board data, phrased by a low-temperature model call and rejected by a validation gate if any number is not in the source row. How the leaderboards work →