Muse Spark Open is the open-weight release of Muse Spark from Meta Superintelligence Labs, an 816B-parameter mixture-of-experts model with 42B active parameters, released on 15 October 2026 under the Muse Model License. This profile collects the published facts a team needs before it downloads, hosts or buys access to the model: architecture, context window, weight formats and hardware, licence conditions, gated access, hosted providers, reasoning settings, provider-reported benchmarks and the safety results in the model card. It is written for developers, platform engineers, buyers and AI governance teams. Every fact comes from Meta's own announcement, developer documentation and Hugging Face model card, listed under Standards and references. Where Meta publishes nothing on a point, the page says so.
Last checked: 2026-11-15.
This is an independent reference page from NHI Mgmt Group based on public provider documentation; it is not sponsored or endorsed by any provider. It uses Meta's public Muse Spark Open documentation as of November 2026. Model details change, so confirm against Meta's current documentation.
Key takeaways
- 816B total, 42B active parameters. A sparse mixture-of-experts transformer with 256 experts, 8 routed per token, and a built-in 1.8B-parameter perception encoder.
- 1,048,576-token context window. Input is text, image, video and audio; output is text. Meta does not state a maximum output or knowledge cutoff on the pages checked.
- No Meta Model API ID and no Meta price. The weights are downloaded and self-hosted, or used through hosted providers such as Baseten, Fireworks AI, Groq, Modal, Nebius, OpenRouter and Together AI, each with its own pricing.
- Muse Model License with a revenue condition. Model-as-a-Service providers above US $20 million in gross revenue over any 12 months must notify Meta and sign a separate agreement, and products must display "Built with Muse Spark Open".
- Gated download. The Hugging Face form asks for full legal name, date of birth and organisation name, cannot be edited after submission, and the model is not intended for users under 18.
What is Muse Spark Open?
Muse Spark Open is the open-weight member of Meta's Muse family. Meta's research post of 15 October 2026 calls it "the open-weight version of Muse Spark" and describes it as natively multimodal and focused on coding and agentic work, including long-horizon tasks, software engineering and multi-step reasoning. Meta's developer documentation calls Muse Spark "Meta's agentic and coding model" and lists the open release for self-hosted or provider-hosted agentic and coding work and for post-training with supervised fine-tuning or reinforcement learning.
Unlike the closed Muse Spark, Muse Spark Open is not served on the Meta Model API: Meta's models page shows no model ID for it. Meta does not state which Muse Spark version (1.1, 1.2 or 1.3) the open weights correspond to. The research post also says the model runs on two Apple M5 Ultra Mac Studios using MLX, which it describes as "the first model in its class to run on your desktop". Meta's other open-weight Muse model is the smaller Muse Glimmer, distilled from Muse Spark. For the hosted model, see the Meta Muse Spark profile; for the full line-up, see Meta Models.
Key facts at a glance
As of 2026-11-15. Source: Meta research post, Muse Spark Open documentation, models page and Hugging Face model card.
| Field | Value | Source |
|---|---|---|
| Provider | Meta (Meta Superintelligence Labs) | Hugging Face model card |
| Repository IDs | meta-models/Muse-Spark-Open-816B-A42B (BF16); meta-models/Muse-Spark-Open-816B-A42B-NVFP4 | Hugging Face model card |
| Meta Model API ID | None (not served on the Model API) | Models page |
| Release date | 2026-10-15 | Research post; model card |
| Status | Current | Models page |
| Context window | 1,048,576 tokens | Muse Spark Open documentation |
| Max output | Not stated by Meta on the pages checked (2026-11-15) | Model card |
| Knowledge cutoff | Not stated by Meta on the pages checked (2026-11-15) | Model card |
| Input modalities | Text, image, video, audio | Models page; model card |
| Output modalities | Text | Models page; model card |
| Open weights | Yes, gated (BF16 and NVFP4) | Model card; documentation |
| Licence | Muse Model License (Muse Spark Open version, dated 15 October 2026) plus Usage Policy | Model card |
| Parameters | 816B total, 42B active (the Hugging Face page header shows 822B) | Model card; documentation |
| Languages | Trained on data from more than 100 languages | Documentation; model card |
Architecture
As of 2026-11-15. Source: Hugging Face model card.
| Component | Value |
|---|---|
| Type | Mixture-of-experts transformer with Perception Encoder |
| Layers and hidden dimension | 62 layers, 8,192 hidden dimension |
| Experts | 256, with 8 activated per token (top-k routing) |
| Attention | Grouped-query attention, 128 query heads, 16 KV heads, head dimension 64; gated attention with QK-norm |
| Vocabulary | 202,048 tokens; input and output embeddings untied |
| Perception Encoder | 1.8B parameters, 50 layers, 1,536 hidden dimension, 4,096 output dimension |
Release and version history
As of 2026-11-15. Source: Meta research post and Hugging Face model card.
| Date | Event | Source |
|---|---|---|
| 2026-10-15 | Muse Spark Open released with weights on Hugging Face and launch partners for hosting, tooling and silicon | Research post; model card |
| 2026-10-15 | Muse Model License (Muse Spark Open version) dated | Model card |
| Not stated | Full evaluation report, described as "available soon" | Research post |
There is one published checkpoint, in two precisions. Meta lists no point releases or revised weights on the pages checked (2026-11-15). The closed Muse Spark it derives from has versions 1.1, 1.2 and 1.3 on the Model API.
How much does Muse Spark Open cost?
As of 2026-11-15. Meta does not sell Muse Spark Open by the token. Cost is either the hardware to run it or a hosted provider's price.
| Item | Price | Notes | Source |
|---|---|---|---|
| Weights download | No fee stated by Meta | Meta describes "limited commercial terms, which allow many developers to use the model for free" | Research post |
| Meta Model API | Not offered | No Model API ID | Models page |
| Hosted providers | Set by each provider | Baseten, Fireworks AI, Groq, Modal, Nebius, OpenRouter, Together AI; Meta does not publish their prices | Documentation; research post |
| Large Model-as-a-Service providers | Separate written agreement with Meta | Required above US $20 million aggregate gross revenue in any consecutive 12 months | Model card (licence) |
| Self-hosting | Operator's hardware and operations | See hardware figures under Context window and output limits | Documentation |
The licence excludes internal use, research and features inside a company's own product from the revenue condition, per the model card. Read the licence text itself before relying on that summary. See the LLM Pricing Tracker.
Context window and output limits
Meta's documentation states a 1,048,576-token context window, and the model card states 1M tokens at inference. Self-hosting at that length is a hardware question: the memory figures below include KV cache only where Meta says so.
| Limit | Value | Notes | Source |
|---|---|---|---|
| Context window | 1,048,576 tokens | 1M tokens at inference | Documentation; model card |
| Max output | Not stated by Meta on the pages checked (2026-11-15) | None | Model card |
| BF16 weights | About 1.65 TB; roughly 2 TB VRAM to serve | 2 TB includes KV cache, activations and runtime overhead; eight GPUs such as 8 x B300 | Documentation |
| NVFP4 weights | About 537 GB; at least 700 GB VRAM | Mostly 4-bit with some BF16 layers; a single 4-GPU NVIDIA Blackwell node | Documentation; model card |
| Desktop option | Two Apple M5 Ultra Mac Studios with MLX | Each with up to 512 GB unified memory | Research post |
| Video input | At least 448 px on the short side recommended | Short or low-resolution videos (about 100 px) may be rejected | Model card |
See Mixture of Experts Explained and Context Windows Explained.
Modalities and tools
- Inputs: text, image, video and audio, handled by the built-in perception encoder; output is text.
- Reasoning effort: set through reasoning_effort or chat_template_kwargs with values low, medium, high or xhigh; xhigh is the default. Meta's deployment page states reasoning cannot be disabled (requests with "none" are rejected), and max_tokens must cover both reasoning and the final answer.
- Recommended sampling: temperature 1.0, top_p 1.0, top_k disabled.
- Tool use: the model card describes sequential tool invocation across multi-turn workflows, including recovery from failed tool calls. Meta's prompting guide covers chat, system prompts, reasoning, tool calling and media input.
- Agentic tests: the research post describes on-device agentic workflows, including iOS changes to the Ice Cubes app and driving DaVinci Resolve.
- Inference engines: SGLang, vLLM, TensorRT-LLM, Hugging Face Transformers, Apple MLX, llama.cpp and Unsloth, with optimisations for AMD, Apple, Intel and NVIDIA silicon.
- Post-training: PyTorch, including TorchTitan; Meta's documentation has a post-training section for SFT and RL.
An open-weight agent model that calls tools carries the credentials of every tool it can reach. Scope each integration, log tool calls and follow Meta's own advice to require human confirmation for irreversible actions. See the Open-Weight Model Security Guide.
Where is Muse Spark Open available?
As of 2026-11-15. Source: Muse Spark Open documentation and research post. Providers listed alphabetically within each row.
| Channel | Details | Source |
|---|---|---|
| Weight downloads | Meta and Hugging Face (gated) | Documentation |
| Model catalogues | Amazon SageMaker AI, Google Cloud Model Garden, Microsoft Foundry Model Catalog | Documentation |
| Hosted inference platforms | Baseten, Fireworks AI, Groq, Modal, Nebius, Together AI | Documentation |
| Gateway | OpenRouter | Documentation |
| Meta Model API, meta.ai, Meta AI app | Not offered for Muse Spark Open on the pages checked (2026-11-15) | Models page |
| Regions | Not stated by Meta on the pages checked (2026-11-15) | Documentation |
Meta names these as launch partners. Each provider sets its own terms, data handling and regions. See Self-Hosted LLMs vs Hosted LLM APIs.
Access restrictions
Muse Spark Open is an open-weight model with gated access. To download from Hugging Face, a user must log in and complete a form with full legal name, date of birth and full organisation name, which cannot be edited after submission. The model card states the model is not intended for users under 18. Use is governed by the Muse Model License and its Exhibit A Usage Policy, which prohibits uses including illegal activity, harassment, unlicensed professional practice, malware, weapons, self-harm, fraud, disinformation and impersonation. The licence requires products to prominently display "Built with Muse Spark Open" in the UI, documentation, website or settings, and sets the separate-agreement condition for large Model-as-a-Service providers described above.
Benchmarks (provider-reported, dated)
As of 2026-11-15. All figures are provider-reported by Meta in the Hugging Face model card (October 2026), at xhigh reasoning effort. The Artificial Analysis Intelligence Index figure is as reported by Meta in the card and was not checked on Artificial Analysis's own pages for this profile.
| Evaluation | Muse Spark Open | Source and date |
|---|---|---|
| Artificial Analysis Intelligence Index | 39 | Model card, 2026-10 |
| MCP-Atlas | 89.2 | Model card, 2026-10 |
| OSWorld Verified | 80.4 | Model card, 2026-10 |
| Terminal-Bench 2.1 | 76.4 | Model card, 2026-10 |
| Toolathlon Verified | 72.8 | Model card, 2026-10 |
| DeepSWE 1.1 | 60 | Model card, 2026-10 |
| SWE-Atlas-QnA | 58 | Model card, 2026-10 |
| GPQA Diamond | 89.0 | Model card, 2026-10 |
| HLE-Diamond (no tools) | 26.1 | Model card, 2026-10 |
| Humanity's Last Exam (with tools) | 61.0 | Model card, 2026-10 |
| ZeroBench (with tools) | 50.0 | Model card, 2026-10 |
| VideoMMEv2 | 61.5 | Model card, 2026-10 |
The card sets these against other open-weight models (GLM 5.3, Kimi K3, Mistral Large 4, Nemotron 3 Ultra, Qwen3.8 and others); those comparator scores are not reproduced here. The research post's "Key Benchmarks" section gives no figures in its text. See How to Read LLM Benchmarks.
Model card and system card
Meta publishes a model card on Hugging Face and refers to an evaluation methodology report; a full evaluation report is "available soon" per the research post. No separate system card is linked on the pages checked (2026-11-15). Meta states the model received "additional safety post-training" because of open-weight considerations and was evaluated under the Meta Superintelligence Scaling Framework, meeting its release thresholds. What the card states:
- Cybersecurity: refused 99.2% of jailbreaks on Meta's internal benchmark and 94.8% across aggregated public benchmarks, without system-level safeguards.
- Chemical and biological: refused over 98% of harmful requests on SecureBio's BioTIER and over 99% on an internal single-turn chemistry benchmark.
- Misalignment: 98.1% honesty on MASK and a 0% misbehaviour rate on Agentic Misalignment.
- Deployment advice: deploy within a broader system with additional guardrails, and require human confirmation for irreversible agentic actions.
- Limitations: may produce inaccurate, biased or objectionable responses and make multi-step reasoning errors, especially in novel scenarios; performance may degrade outside strongly supported languages; developers are responsible for their own safety testing.
See Model Cards and System Cards Explained and the Frontier AI Model Risk Frameworks Guide.
Data handling
- Self-hosted weights: prompts and outputs stay in the operator's environment; Meta states no inference data flow for downloaded weights on the pages checked.
- Hosted providers: retention, training use and residency are set by each provider, not by Meta.
- Download form: the gated Hugging Face form collects legal name, date of birth and organisation; the Meta download route requires sign-in and licence acceptance, after which the model runs offline with "no API key", per Meta's get the model page.
- Training data: not described by Meta on the pages checked (2026-11-15), beyond more than 100 languages.
Self-hosting notes from Meta's deployment guide
- Network exposure: several example server configurations bind to 0.0.0.0; TensorRT-LLM defaults to localhost.
- Authentication: the llama.cpp API key is optional and its /health and /v1/models endpoints answer without authentication; the Unsloth API requires a bearer key.
- Remote code: the recommended TensorRT-LLM configuration sets trust_remote_code to true.
- Integrity: the BF16 download has 65 weight shards and NVFP4 127; Meta warns a runtime cannot load the checkpoint if any file is missing or renamed, and the page describes no checksums, so record your own hashes.
- Served context: TensorRT-LLM's validated serving window is 49,152 tokens, well below the 1,048,576-token maximum.
Self-hosting moves patching, access control and logging to the operator. See the Self-Hosted LLM Security Guide.
Deprecation status
As of 15 November 2026, Meta's documentation, models page and model card list no deprecation, end-of-support date or replacement for Muse Spark Open. Downloaded weights do not expire; hosted providers may retire their endpoints on their own timelines. Monitor Meta's developer documentation rather than relying on this snapshot.
| Model or item | Details | Source |
|---|---|---|
| Muse-Spark-Open-816B-A42B (BF16, NVFP4) | No deprecation listed (2026-11-15) | Model card; models page |
| Hosted provider endpoints | Set by each provider; not stated by Meta | Documentation |
How Muse Spark Open differs from other models in its family
As of 2026-11-15. Source: Meta models page, Muse Spark Open documentation and model card. Models listed alphabetically.
| Model | API ID | Weights | Context | Input | Positioning (Meta wording) |
|---|---|---|---|---|---|
| Muse Ember | muse-ember-1.0 | Closed | Not stated on the models page (2026-11-15) | Text, image, video, audio, PDF | Efficient model for high-volume work such as sub-agent tasks and tool calls |
| Muse Glimmer | None | Open | Not stated on the models page (2026-11-15) | Text, image | Open-weight, self-hosted, distilled from Muse Spark |
| Muse Spark | muse-spark-1.3, 1.2, 1.1 | Closed | 1,048,576 | Text, image, video, audio, PDF | Agentic and coding work at scale |
| Muse Spark Open | None | Open (gated) | 1,048,576 | Text, image, video, audio | Open-weight, self-hosted or provider-hosted |
| Muse Wonder | muse-wonder-1.0 | Closed | 1,048,576 | Text, image, video, audio, PDF | Complex software engineering and long-horizon agentic work; in preview |
On Meta's models page, the open model lists text, image, video and audio input, while the hosted Muse models also list PDF. Its reasoning effort runs from low to xhigh and cannot be switched off. For other open-weight models, see Open-Weight Frontier Models Compared.
Practitioner checklist
- Have legal review the Muse Model License, the Usage Policy and the "Built with Muse Spark Open" attribution duty before any product use.
- If you resell inference, track gross revenue against the US $20 million condition that triggers a separate agreement with Meta.
- Decide who in the organisation completes the gated form, since the details cannot be edited after submission.
- Size hardware from Meta's figures: about 700 GB VRAM for NVFP4 and roughly 2 TB for BF16, plus headroom for long contexts.
- Download only from Meta or the official Hugging Face repositories, keep original file names and record your own file hashes.
- Bind self-hosted inference servers to localhost or a private network and require authentication on every endpoint.
- For hosted use, review each provider's retention, training-use and region terms, as Meta does not set them.
- Add your own guardrails and require human confirmation for irreversible tool actions, as the model card advises.
- Store provider API keys in a secrets manager and set spend limits. See the LLMjacking Guide.
- Recheck Meta's documentation for the full evaluation report and any revised weights.
Standards and references
- Meta: Muse Spark Open model card on Hugging Face (accessed 2026-11-15)
- Meta: Muse Spark Open documentation (accessed 2026-11-15)
- Meta: Model API models (accessed 2026-11-15)
- Meta: Get Muse Spark Open (accessed 2026-11-15)
- Meta: Run Muse Spark Open inference (accessed 2026-11-15)
- Meta AI: Introducing Muse Spark Open (accessed 2026-11-15)
Related NHI Mgmt Group resources: Meta Models, Meta Muse Spark, Open-Weight Frontier Models Compared, Open-Weight Model Security Guide, Self-Hosted LLM Security Guide.