Muse Spark
- Provider
- Meta
- Status
- available
- Context
- 262,000 tok
- SWE-bench
- 55%
Muse Spark is the first model from Meta’s Meta Superintelligence Labs, launched on 8 April 2026 and spearheaded by chief AI officer Alexandr Wang. It is the clearest signal of Meta’s strategy reset: a proprietary, natively multimodal reasoning model that replaced the open Llama models as the engine behind Meta AI — a deliberate break from the open-weights approach Meta built its AI reputation on (CNBC).
Meta describes it as “small and fast by design,” and the efficiency claim is the headline: Meta says Muse Spark matches Llama 4 Maverick with over 10x less compute (Artificial Analysis). It is a capable first-generation frontier-adjacent model — independent evaluator Artificial Analysis scores it 52 on its Intelligence Index v4.0, 4th overall at release — that leads on health questions and vision, but sits below the very top of the field.
Quick specs
| Provider | Meta (Meta Superintelligence Labs) |
| Released | 8 April 2026 |
| Status | Available (free in Meta AI; API private preview) |
| Context window | ~262,000 tokens |
| Modalities | Text, image and speech in; text out |
| AA Intelligence Index | 52 (4th overall at release) |
| MMMU-Pro (vision) | 80.5% |
| HealthBench Hard | 42.8 (category leader) |
| Licence | Proprietary (weights not released) |
| Best for | Everyday multimodal assistance, health questions, vision, efficiency |
| Limitations | Below the frontier leaders; proprietary; API private preview |
What Muse Spark is
Muse Spark is the first in a planned “Muse” family from Meta Superintelligence Labs, the organisation Mark Zuckerberg built around Alexandr Wang after a $14 billion-plus deal for a stake in Scale AI (CNBC). It is natively multimodal — accepting text, image and speech input — with tool use, visual chain-of-thought and an Instant / Thinking mode split for quick answers versus deeper reasoning (Meta).
The strategically important fact is that it is proprietary. Where Llama defined Meta’s open-weights era, Muse Spark keeps its weights closed (Meta says it “hopes to open-source future versions”), prioritising integration into Meta’s own products and ecosystem (Artificial Analysis). See the Meta provider page for the wider pivot.
Benchmark performance
Vendor- and Artificial-Analysis-reported figures; treat as a ceiling pending broader independent testing.
| Benchmark | Muse Spark | Notes |
|---|---|---|
| AA Intelligence Index v4.0 | 52 | 4th overall (Gemini 3.1 Pro 57, GPT-5.4 57, Claude Opus 4.6 53) |
| MMMU-Pro (vision) | 80.5% | 2nd-most-capable vision model at release |
| HealthBench Hard | 42.8 | Leads GPT-5.4 (40.1), Gemini 3.1 Pro (20.6), Grok 4.2 (20.3) |
| Humanity’s Last Exam (Contemplating) | ~58% | With parallel multi-agent reasoning |
Muse Spark’s profile is “strong generalist, standout on health and vision, efficient” rather than frontier-leading (Artificial Analysis). Its Contemplating mode orchestrates multiple agents reasoning in parallel, letting it compete with the extreme-reasoning modes of frontier models (≈58% on Humanity’s Last Exam). On our best AI models ranking it sits in the Strong tier.
Capabilities
Muse Spark handles text, image and speech input with text output, a ~262K-token context window, tool use, visual chain-of-thought and multi-agent orchestration (the Contemplating mode). The efficiency story is central: Meta claims it reaches Llama 4 Maverick-level capability with more than ten times less compute (Artificial Analysis), consistent with the “small and fast by design” framing.
Access and pricing
Muse Spark is free for general users on meta.ai and the Meta AI app, where it powers the assistant across Meta’s products (Meta). It is proprietary, offered only in private preview via API to select partners, with no public per-token price. There is no open-weights release — a notable contrast with Llama 4.
How Muse Spark compares
- vs the frontier — Muse Spark trails Claude Opus 4.8, GPT-5.5 and Gemini 3.1 Pro on the aggregate Intelligence Index, but leads them on HealthBench and is near the top on vision.
- vs Llama 4 — Muse Spark replaced Llama as the Meta AI engine and matches Llama 4 Maverick’s capability at a fraction of the compute, but it is closed where Llama is open.
Known limitations
Below the frontier leaders on overall capability. Proprietary — no open weights, a reversal of Meta’s Llama strategy and a disappointment to the open-source community. API is private preview only, so developers can’t yet build on it freely. And the wider Meta AI privacy posture (ad personalisation from chats) applies — see Meta AI.
FAQ
What is Muse Spark?
Muse Spark is the first model from Meta Superintelligence Labs, launched 8 April 2026. It is a proprietary, natively multimodal reasoning model that replaced Llama as the engine behind Meta AI, and Meta’s first major step away from open weights.
How good is Muse Spark?
Capable but not frontier-leading. Artificial Analysis scores it 52 on its Intelligence Index (4th overall at release), behind Gemini 3.1 Pro, GPT-5.4 and Claude Opus 4.6 — but it leads on HealthBench Hard and is among the best vision models, and is notably compute-efficient.
Is Muse Spark open source?
No. Unlike Llama, Muse Spark is proprietary — Meta has not released the weights, though it says it hopes to open-source future versions. It is free to use in Meta’s apps and in private API preview.
How much does Muse Spark cost?
It is free for consumers in the Meta AI app and at meta.ai. There is no public per-token API price; access is private preview only.
Last verified 18 June 2026. Figures are Meta- and Artificial-Analysis-reported from the April 2026 launch. Confirm against Meta’s official pages before relying on specific numbers.