Meta · model passport
Muse Spark
muse-spark- Cloud only
- Proprietary licence
- Released 2026-09-13(first seen on LMArena)
- Overall · LMArena
- 1,489#13 of 396
- Best percentile
- 97.0Overall
- Cohort standing
- 79.2median 60.4
What the data says
Benchmarked on 12 of 27 tracked axes. Every figure below is the model's own published result on that board.
Strongest axes
- 97 pct — Overall (1,489)
- 96 pct — Vision Arena (1,294)
- 95.5 pct — Hard prompts (1,506)
Uneven: 36.5 percentile points between its best and worst axis — 97 pct on Overall versus 60.5 pct on Document Arena.
- Overall97 pct1,489Verify →
- Vision Arena96 pct1,294Verify →
- Hard prompts95.5 pct1,506Verify →
- Coding95.1 pct1,529Verify →
- IFBench (Artificial Analysis)93.5 pct75.9%Verify →
- Instruction following90.4 pct1,464Verify →
- FrontierMath88.4 pct39.0%Verify →
- Humanity's Last Exam87 pct40.6%Verify →
- Factuality Arena83.7 pct1,468Verify →
- GPQA Diamond80.1 pct89.8%Verify →
- AIME (math)71 pct88.9%Verify →
- Document Arena60.5 pct1,467Verify →
- WebDev ArenaNot benchmarked
- Search ArenaNot benchmarked
- Agent ArenaNot benchmarked
- SimpleQA VerifiedNot benchmarked
- ARC-AGI-2Not 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
- 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.
Muse Spark stands at 79.2 (raw mean 86.5) — measured on 12 of 27 axes.
Muse Spark stands at 79.2 versus a cohort median of 60.4 across 175 models released within 180 days — above the peer median.
Muse Spark scores in the 97th percentile on Overall.
Muse Spark scores in the 96th percentile on Vision Arena.
Muse Spark ranks between #16 and #40 across 5 built-in weightings — best under "All sources", worst under "Coding-focused".
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 →