Muse Spark: benchmarks, price and rank
Muse Spark from Meta is 59th of 123 rated AI models on Spezia's leaderboard.
- 59th of 123 rated models
- Rating 1,249
- Rated on 7 benchmarks
- Closed
- Released Apr 8, 2026
Where it is strongest
Not in the top tenth on any benchmark in use.
Where it falls behind
- 7th of 8 on DeepSearchQA
- 42nd of 50 on SWE-Bench Pro
- 51st of 94 on Epoch Capabilities Index
Its headline results
The benchmarks that count most in the rating, with its place among the models measured the same way, and where each number came from.
- SWE-Bench Pro52.4%
Long changes to real codebases in several languages, often across many files, including code from private and copyleft projects that models are unlikely to have seen.
42nd of 50. The maker's own number, from Meta, via LLM Stats.
- Humanity's Last Exam40.6%
About 2,500 questions written by experts in more than a hundred fields to be as hard as they could make them, many needing a diagram or image.
Independent result from Epoch AI.
- MMMU-Pro80.4%
College exam questions that need a picture to answer, such as charts, diagrams, maps and chemical structures, with more answer options than the original MMMU and some questions shown only as a screenshot.
14th of 51. The maker's own number, from Meta, via LLM Stats.
- CharXiv Reasoning86.4%
Questions about real charts from scientific papers that need reasoning across the chart, not just reading off a value.
14th of 45. The maker's own number, from Meta, via LLM Stats.
- ScreenSpot-Pro84.1%
Screenshots of professional software at full resolution, with an instruction to find the exact button or field to use.
5th of 18. The maker's own number, from Meta, via LLM Stats.
- LMArena text1,474
People put a question to two anonymous models and vote for the better answer.
22nd of 107. Independent result from LMArena.
- ERQA64.7%
400 questions about images of the physical world: space, motion, actions, the state of objects and views from several cameras.
11th of 21. The maker's own number, from Meta, via LLM Stats.
- DeepSearchQA74.8%
Questions that need many searches and several steps of reasoning across sources to answer.
7th of 8. The maker's own number, from Meta, via LLM Stats.
- SWE-bench Verified77.4%
Real bug reports and feature requests from popular open-source Python projects. The model has to change the code so the project's hidden tests pass.
11th of 56. The maker's own number, from Meta, via LLM Stats.
- GPQA Diamond89.8%
Graduate-level questions in biology, physics and chemistry, written by PhDs so that searching the web does not give the answer away.
48th of 103. Independent result from Epoch AI.
3 more results in its full profile.
Every result with its source, what it costs per task, what jumped with its release, side by side with any other model, and answers to your questions about when to use it.
Models ranked near it
How the rating works, and every benchmark it uses: the model leaderboard.