Muse Spark 1.3

MetaReleased 2 Sep 2026ClosedFlagship

Compare

170159–182Index score#17 · provisional, limited data

2/ 8Core tests taken2 of 6 areas · plus 6 anchor tests

−22Versus the frontier at releasebest then: Claude Fable 5.1

$1.25/ $4.25Price per million tokensinput / output, list price

1.05MContext windowtokens it reads at once

Where it sitson the frontier

GPT-5.4 reached this score 6 months earlier.

Test resultsby area · tick = best by any model

Reasoning & knowledgeno evidence yet

GPQA Diamond

no result

Humanity's Last Exam

no result

Mathsevidence

FrontierMath T1-3

74%

Codingno evidence yet

SWE-bench Verified

no result

Terminal-Bench

no result

Agentsevidence

APEX-Agents

58%

Novel problemsno evidence yet

ARC-AGI-2

no result

Long tasksno evidence yet

METR time horizon

no result

Independent results from the Epoch AI benchmarking hub, best reasoning setting. The core tests are shown here; anchor tests also inform the score and can be shown above.

What others sayrank on each leaderboard

The Model Indexindependent tests, our method#17of 219 · 170

LMArena · Mathsmaths questions in chat#7of 373 · 1512 LMArena · Textoverall chat quality, judged by people#8of 388 · 1495 LMArena · Codingcoding questions in chat#10of 383 · 1539 Artificial Analysis Intelligence Indexoverall capability across ten evaluations#7of 207 · 48 LMArena · Creative writingcreative writing#25of 386 · 1459 LMArena · Visionunderstanding images#10of 149 · 1290 Epoch Capabilities Indexoverall capability from many benchmarks#17of 239 · 156.9 LMArena · WebDevbuilding web apps and front ends#12of 123 · 1655 LiveBench · Overalloverall, with fresh questions to limit contamination#6of 58 · 81.6 LMArena · Documentsworking with long documents#14of 41 · 1471

Further left is better. Leaderboards measure different things: LMArena is people voting blind between two answers, the others are test scores. Click any row for the source.