Skip to content
aiqbench

Mistral · model passport

Mistral Medium 2508

mistral-medium-2508
  • Cloud only
  • Proprietary licence
  • Released 2025-08-01(from model name)
7/27axes covered
Overall · LMArena
1,408#141 of 396
Best percentile
65.4Hard prompts
Cohort standing
48.5median 51.6

What the data says

Benchmarked on 7 of 27 tracked axes. Every figure below is the model's own published result on that board.

Weakest axes

  • 3.8 pct — HHEM Hallucination (Vectara) (77.3%)

Uneven: 61.6 percentile points between its best and worst axis — 65.4 pct on Hard prompts versus 3.8 pct on HHEM Hallucination (Vectara).

  • Hard prompts65.4 pct1,427Verify →
  • Coding64.7 pct1,453Verify →
  • Overall64.6 pct1,408Verify →
  • Instruction following64.1 pct1,395Verify →
  • Factuality Arena37.8 pct1,429Verify →
  • Vision Arena34.7 pct1,159Verify →
  • HHEM Hallucination (Vectara)3.8 pct77.3%Verify →
  • WebDev ArenaNot benchmarked
  • Search ArenaNot benchmarked
  • Document ArenaNot benchmarked
  • Agent ArenaNot benchmarked
  • GPQA DiamondNot benchmarked
  • AIME (math)Not benchmarked
  • FrontierMathNot benchmarked
  • SimpleQA VerifiedNot benchmarked
  • ARC-AGI-2Not benchmarked
  • Humanity's Last ExamNot benchmarked
  • SimpleBenchNot benchmarked
  • LiveBenchNot benchmarked
  • BFCL v4 (function calling)Not 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
  • 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-2508 stands at 48.5 (raw mean 47.9) — measured on 7 of 27 axes.

    standingtemplateVerify →

  • mistral-medium-2508 stands at 48.5 versus a cohort median of 51.6 across 144 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 →