Mistral · model passport
Magistral Medium 2506
magistral-medium-2506- Cloud only
- Proprietary licence
- Released 2025-06-01(from model name)
- Overall · LMArena
- 1,305#261 of 395
- Best percentile
- 47.7Coding
- Cohort standing
- 40.5median 50.8
What the data says
Benchmarked on 5 of 27 tracked axes. Every figure below is the model's own published result on that board.
Weakest axes
- 11 pct — ARC-AGI-2 (0.0%)
Uneven: 36.7 percentile points between its best and worst axis — 47.7 pct on Coding versus 11 pct on ARC-AGI-2.
- Coding47.7 pct1,387Verify →
- Hard prompts42 pct1,333Verify →
- Instruction following39.2 pct1,301Verify →
- Overall34.2 pct1,305Verify →
- ARC-AGI-211 pct0.0%Verify →
- WebDev ArenaNot benchmarked
- Vision ArenaNot benchmarked
- Factuality ArenaNot benchmarked
- Search ArenaNot benchmarked
- Document ArenaNot benchmarked
- Agent ArenaNot benchmarked
- GPQA DiamondNot benchmarked
- AIME (math)Not benchmarked
- FrontierMathNot benchmarked
- SimpleQA VerifiedNot 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
- 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.
The magistral-medium-2506 model from mistral covers 5 of 27 axes with a composite shrinkage of 40.5 and a raw composite score of 34.8, without thin coverage.
standingVerify →The magistral-medium-2506 model from mistral has a standing of 40.5, which is below the cohort median standing of 50.8 across 142 models in an 180-day window.
cohortVerify →
magistral-medium-2506 from mistral scored in the 11th percentile on ARC-AGI-2.
weaknessVerify →
The magistral-medium-2506 model from mistral shows a sensitivity spread of 618 between the Coding-focused and All sources templates across 5 covered templates, with ranks ranging from 181 to 799.
rank swingVerify →
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 →