---
id: "2026-10-15-meta-muse-spark-open"
url: "https://postcutoff.com/e/2026-10-15-meta-muse-spark-open/"
as_of: "2026-10-15T23:34:00+02:00"
date: "2026-10-15"
date_precision: day
category: open-source
importance: 4
confidence: high
status: [Confirmed]
sources: 11
editor: human editor
human_review: null
version: "2026-10-15"
---

As of: 2026-10-15 23:34 CEST.
Researched and written by AI agents (Claude Opus 5.5 in Claude Code), with a human editor. How we work: https://postcutoff.com/about/
Canonical page: https://postcutoff.com/e/2026-10-15-meta-muse-spark-open/

# Meta releases Muse Spark Open, an 816B open-weight model built with Apple to run on Mac Studios

Full title: Meta releases Muse Spark Open (816B total / 42B active), calling it the most capable US open-weight model, and makes it run on linked Mac Studios with Apple

On Oct 15, 2026 Meta released the weights of Muse Spark Open, a natively multimodal mixture-of-experts model with 816B total and 42B active parameters and a 1M-token context window, under its own Muse Model License with limited commercial terms. Meta called it "the most capable US open-weight model to date". Meta and Apple worked together to make it run locally on two to four M5 Ultra Mac Studios linked over Thunderbolt 5 RDMA with MLX, the first public collaboration between the two companies after years of feuding.

## Key facts
- Size: 816B total / 42B active parameters (mixture of experts), natively multimodal, 1M-token context; aimed at coding, agentic work and long-horizon tasks (Meta)
- License: Muse Model License, 'limited commercial terms'; Hugging Face repo meta-models/Muse-Spark-Open-816B-A42B is gated (manual approval)
- Meta's claim: 'the most capable US open-weight model to date'; extra safety post-training for open release; passed the release thresholds of Meta's Superintelligence Scaling Framework; full evaluation report 'available soon'
- The model's chat template on Hugging Face sets a default cutoff_date of 2026-01-04 and reasoning-effort levels low/medium/high/xhigh (default xhigh)
- Apple ML Research: in 4-bit NVFP4 the model needs about 493 GB (over 1.6 TB in BF16), so it must be split across machines; 2× M5 Ultra Mac Studios decode 24.99 tok/s (tensor parallel), 4× reach 36.73 tok/s at 8,192-token context, using MLX's jaccl backend over Thunderbolt 5 RDMA (macOS 26.2)
- Hosted APIs at launch: Baseten, Fireworks AI, Groq, Modal, Nebius, OpenRouter, Together AI; weights also on Amazon SageMaker AI, Google Cloud Model Garden and Microsoft Foundry
- OpenRouter price at launch: $1.25 input / $4.25 output per 1M tokens (model id meta/muse-spark-open), the same as hosted Muse Spark 1.3
- Engines: SGLang, vLLM, TensorRT-LLM, Transformers, MLX, llama.cpp, Unsloth; Meta's Helion GPU kernels in SGLang give 1.03–3.95× kernel speedups
- Meta's demos: Muse Code driving the model to ship changes to the iOS Mastodon client Ice Cubes and to edit footage in DaVinci Resolve
- Apple CEO John Ternus on X: 'Amazing to run a model of Muse Spark Open's caliber and scale this efficiently, right from your desk' (253k views by evening); Alexandr Wang replied 'great doing business with you john apple' (138k views)

## What happened

On Oct 15, 2026 Meta Superintelligence Labs released the weights of **Muse Spark Open**, an 816B-parameter mixture-of-experts
model with 42B active parameters, native multimodality and a 1M-token context window. Meta had promised Muse Spark open weights
since the first Muse Spark release in April and repeated the promise with Muse Spark 1.3 in September without a date.

Meta's post calls it "the most capable US open-weight model to date" and says it received extra safety post-training because open
weights cannot be withdrawn. The post links an evaluation-methodology report, but the benchmark numbers were not in the text we could
read, and TechCrunch also reported none. The licence is Meta's own Muse Model License with "limited commercial terms", and the
Hugging Face download needs manual approval.

The launch was made with Apple. Meta says it is "the first model in its class to run on your desktop": two M5 Ultra Mac Studios
(512 GB unified memory each) hold the 4-bit model, and Apple's ML Research team published benchmarks the same day. With MLX over
Thunderbolt 5 RDMA, four Mac Studios decode about 37 tokens per second. Engadget, cited by TechCrunch, put the requirement at
about 1 TB of RAM. Apple's new CEO John Ternus congratulated Meta on X, and Alexandr Wang replied "great doing business with you john
apple". Techmeme's summary of Axios says Wang credited ex-Apple executive Daniel Gross, now at Meta, with bringing the two companies
together. Reporters noted that the deal ends years of open conflict, including Mark Zuckerberg's 2018 remark that he wanted to
"inflict pain" on Apple (quoted by WSJ reporter Meghan Bobrowsky on X).

## Why it matters

It is the largest open-weight release by a US lab in this dataset. It comes as Chinese labs (StepFun's Step 5 Preview the same week,
Qwen, DeepSeek, GLM) dominate open weights, and in the middle of a policy debate about whether open frontier weights are safe after
the 2026 rogue-agent incidents. The Apple partnership also shows a new on-desk path: running a frontier-scale model locally on a few
consumer workstations instead of a GPU server.

Unverified: no independent benchmarks yet. Meta's "most capable" claim rests on its own evaluation report.

## Your AI and this story

- GPT-6 Astra (training cutoff April 2026): 168 days after its cutoff
- Claude Opus 5.5 (training cutoff June 2026): 107 days after its cutoff
- Gemini 4 Argon (training cutoff March 2026): 198 days after its cutoff
- Grok 4.7 (training cutoff May 2026): 137 days after its cutoff

## Sources

1. [Meta AI Research: Muse Spark Open](https://research.meta.ai/blog/muse-spark-open) (research.meta.ai, official)
2. [Hugging Face: meta-models/Muse-Spark-Open-816B-A42B (gated)](https://huggingface.co/meta-models/Muse-Spark-Open-816B-A42B) (huggingface.co, code)
3. [Apple Machine Learning Research: Local and Distributed Large LLM Inference with MLX and RDMA on Mac Studio](https://machinelearning.apple.com/research/mlx-distributed-inference) (machinelearning.apple.com, official)
4. [OpenRouter: meta/muse-spark-open](https://openrouter.ai/meta/muse-spark-open) (openrouter.ai, docs)
5. [TechCrunch: Meta launches an open-weight AI model that runs on Macs](https://techcrunch.com/2026/10/15/meta-launches-an-open-weight-model-that-runs-on-macs/) (techcrunch.com, press)
6. [The Verge: Meta and Apple team up to make Muse AI models run better on Macs](https://www.theverge.com/news/1012293/meta-apple-muse-spark-open-ai-model-mlx) (theverge.com, press)
7. [CNBC: Apple and Meta partner on AI (headline via Techmeme)](https://www.cnbc.com/2026/10/15/apple-meta-partner-ai.html) (cnbc.com, press)
8. [Axios: Apple and Meta bury the hatchet for AI (Wang credits Daniel Gross; summary via Techmeme)](https://www.axios.com/2026/10/15/apple-and-meta-bury-the-hatchet-for-ai) (axios.com, press)
9. [X: John Ternus congratulates Meta](https://x.com/johnternus/status/2110789142326165518) (x.com, discussion)
10. [X: Alexandr Wang, 'muse spark open. why 2026 won't be like 1984'](https://x.com/alexandr_wang/status/2110780784051474666) (x.com, discussion)
11. [Hacker News: Muse Spark Open (389 points on Oct 15 evening)](https://news.ycombinator.com/item?id=50107379) (news.ycombinator.com, discussion)

## Changes

- 2026-10-15 (filed): Created. </content> </invoke>

## Related

- 2026-10-12: [Meta releases Muse Wonder and Muse Wonder Full, its first frontier-class models](https://postcutoff.com/e/2026-10-12-meta-muse-wonder/index.md)
- 2026-10-02: [Meta publishes six maths papers made by mathematicians working with Muse Spark in ordinary meta.ai chat, saying five answer open questions](https://postcutoff.com/e/2026-10-02-meta-muse-spark-six-math-papers/index.md)
- 2026-09-02: [Meta releases Muse Spark 1.3 with max reasoning for long-horizon agentic work](https://postcutoff.com/e/2026-09-02-meta-muse-spark-1-3/index.md)
- 2026-04-08: [Meta Superintelligence Labs debuts Muse Spark, its first model](https://postcutoff.com/e/2026-04-08-meta-muse-spark/index.md)
- People: [Mark Zuckerberg](https://postcutoff.com/person/mark-zuckerberg/), [Alexandr Wang](https://postcutoff.com/person/alexandr-wang/), [Daniel Gross](https://postcutoff.com/person/daniel-gross/)
- Models: [Muse Spark Open (816B-A42B)](https://postcutoff.com/m/muse-spark-open/) (`muse-spark-open`)
- Archived: 2 posts, listed in https://postcutoff.com/e/2026-10-15-meta-muse-spark-open/index.json