Mistral · model passport
Mistral 7b Instruct
mistral-7b-instruct- Open weights
- Apache 2.0 licence
- Released 2023-09-27(curated)
- Overall · LMArena
- 1,110#379 of 396
- Best percentile
- 5.6Hard prompts
- Cohort standing
- 18.0median 26.5
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
- 2.6 pct — MMLU-Pro (Artificial Analysis) (24.5%)
- 3.1 pct — IFBench (Artificial Analysis) (19.9%)
- 3.8 pct — LiveCodeBench (Artificial Analysis) (4.6%)
Uneven: 3 percentile points between its best and worst axis — 5.6 pct on Hard prompts versus 2.6 pct on MMLU-Pro (Artificial Analysis).
- Hard prompts5.6 pct1,115Verify →
- Instruction following5.3 pct1,087Verify →
- Coding5.1 pct1,145Verify →
- Overall4.5 pct1,110Verify →
- LiveCodeBench (Artificial Analysis)3.8 pct4.6%Verify →
- IFBench (Artificial Analysis)3.1 pct19.9%Verify →
- MMLU-Pro (Artificial Analysis)2.6 pct24.5%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
- 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
- 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-7b-instruct stands at 18 (raw mean 4.3) — measured on 7 of 27 axes.
mistral-7b-instruct stands at 18 versus a cohort median of 26.5 across 62 models released within 180 days — below the peer median.
mistral-7b-instruct scores at percentile 2.6 on MMLU-Pro (Artificial Analysis).
mistral-7b-instruct scores at percentile 3.1 on IFBench (Artificial Analysis).
mistral-7b-instruct ranks between #394 and #1,032 across 5 built-in weightings — best under "Conversation-first", worst under "All sources".
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 →