Muse Spark

Released

Muse Spark is an AI model released by Meta on Wednesday, Apr 8 2026, 368 days after LLaMA 4 Maverick. Benchmark results (shown below) cover Gray Swan IPI, SWE-Bench Pro, SWE-Bench Verified, DeepSWE 1.1, Terminal-Bench 2.1, MCP Atlas, and 9 more.

Benchmarks

Coding

SWE-Bench ProAgentic coding — Can the AI fix real bugs in real software? It's handed actual problems from open-source projects and has to write code that genuinely solves them. Higher is better.
55%
#32 of 47
SWE-Bench VerifiedCoding — Real coding tasks pulled from open-source projects — the AI has to find and fix actual bugs. A human-checked version of the original SWE-Bench. Higher is better.
77.4%
#24 of 95
DeepSWE 1.1Agentic coding — Artificial Analysis' independent test of deep, agentic software-engineering work — the AI has to plan and carry out substantial coding tasks end to end. (Version 1.1 of the test.) Higher is better.
10%
#43 of 43

Terminal & CLI

Terminal-Bench 2.1Agentic terminal coding — Can the AI work in a command-line terminal — running commands and finishing technical setup tasks the way a developer would? Higher is better.
67.3%
#31 of 44

Agentic & tool use

MCP AtlasMulti-step tool use — Can the AI chain together many tools and steps to complete one bigger task, rather than doing just a single thing? Higher is better.
82.2%
#6 of 19
JobBenchProfessional tool use — Tests the AI on professional workplace tasks that require using real work tools — the kind of multi-step jobs an office worker handles. Higher is better.
17%
#13 of 13
Toolathlon-VerifiedPersonal tool use — Tests how well the AI uses everyday personal tools and apps to get things done — a human-checked version of Toolathlon. Higher is better.
49.4%
#10 of 10
OSWorld-VerifiedAgentic computer use — Can the AI actually operate a computer — clicking, typing, and using real apps — to finish tasks on its own? Higher is better.
53.3%
#31 of 36

Reasoning & science

Humanity's Last ExamMultidisciplinary reasoning — Humanity's Last Exam — extremely hard expert questions across many subjects. “With tools” means the AI is allowed to search the web or run code while answering. Higher is better.
50.4%with tools
#33 of 55
ARC-AGI-2Abstract reasoning — Puzzle-style tests of abstract reasoning and pattern-finding — the kind of thing people find easy but AIs often struggle with. Higher is better.via BenchLM
42.5%
#36 of 42
GPQA DiamondScience — Graduate-level science questions in biology, physics, and chemistry — hard enough that subject-matter PhDs score around 65%. Higher is better.
89.5%
#31 of 145

Multimodal

CharXiv ReasoningChart reasoning — Can the AI read and reason about complex charts and figures, not just text? Higher is better.
88.9%
#2 of 32
BabyVisionVisual reasoning — Tests core visual reasoning — seeing and understanding images the way even young children can, which AIs often find surprisingly hard. Higher is better.
39.9%
#4 of 5
MMMU-ProMultimodal reasoning — A tougher version of MMMU — college-level questions that mix images, diagrams, and text together. Higher is better.
80.4%
#14 of 36

Robustness

Gray Swan IPIPrompt injection robustness — Attackers hide malicious instructions inside content the AI reads — a web page, an email, a document — and try to hijack what it does. Gray Swan's indirect prompt injection benchmark measures how often such an attack succeeds when the attacker gets a single try. Lower is better.
2.9%k = 1
14.3%k = 10
16.5%k = 15
k = 1
#5 of 13
k = 10
#5 of 13
k = 15
#7 of 15

Source: BenchLM, retrieved 11 July 2026. Other sources are identified on the linked benchmark pages.

Compare Muse Spark with

Muse Spark

About

Muse Spark was introduced on April 8, 2026 as the first Muse-family model from Meta Superintelligence Labs. Meta described it as a natively multimodal reasoning model with tool use, visual reasoning and multi-agent orchestration. It launched in Meta AI, with a private API preview for selected users.

Meta presented the release as the result of rebuilding its model-training stack, with further scaling planned. The announcement also described Contemplating mode, which used multiple agents reasoning in parallel and was scheduled for gradual rollout. That product mode was distinct from an ordinary single-model response, and the launch did not establish that Meta had permanently retired the Llama name.

Frequently asked questions

Muse Spark was released by Meta on Wednesday, Apr 8 2026.

All Meta releases

19 tracked