# Muse Spark Open 816B A42B [Canonical model page](https://headlessthrones.com/models/muse-spark-open-816b-a42b) · [Model library](https://headlessthrones.com/models) Public catalog reference from Headless Thrones. Exact catalog ID: `muse-spark-open-816b-a42b`. Catalog source-check date: 2026-10-23. This brief contains the public catalog answers and any catalog guide, not private drafts, historical editorial revisions or independently measured rankings. The canonical HTML page may also display separately published editorial. Current catalog answers ## What to know about Muse Spark Open 816B A42B AI-compiled from linked model documentation. This is a catalog reference, not an independent performance review. Catalog checked October 23, 2026. [Exact model](https://headlessthrones.com/models/muse-spark-open-816b-a42b#model-answer-identity) · [Access](https://headlessthrones.com/models/muse-spark-open-816b-a42b#model-answer-access) · [Token limits](https://headlessthrones.com/models/muse-spark-open-816b-a42b#model-answer-context) · [Inputs & outputs](https://headlessthrones.com/models/muse-spark-open-816b-a42b#model-answer-modalities) · [License](https://headlessthrones.com/models/muse-spark-open-816b-a42b#model-answer-license) · [Model details](https://headlessthrones.com/models/muse-spark-open-816b-a42b#model-answer-research) ### Which Muse Spark Open 816B A42B does this page describe? Muse Spark Open 816B A42B is a model record from **Meta** in the Muse Spark family. [Source](https://dev.meta.ai/docs/muse-spark-open) The recorded model identifier or checkpoint is `meta-models/Muse-Spark-Open-816B-A42B`. Use that exact identifier when checking documentation or a provider listing. [Source](https://dev.meta.ai/docs/muse-spark-open) ### How can you access Muse Spark Open 816B A42B? The catalog documents **Open weights** access for Muse Spark Open 816B A42B. [Source](https://dev.meta.ai/docs/muse-spark-open) As of the catalog check, this version was listed as available. [Source](https://dev.meta.ai/docs/muse-spark-open) Open weights means the model weights are distributed for use under their stated terms. It does not establish a hosted API, an open-source license, or a hardware requirement. ### What are Muse Spark Open 816B A42B's context and output limits? The documented **context window is 1,048,576 tokens**. [Source](https://dev.meta.ai/docs/muse-spark-open) A separate maximum-output value is not established in this record. The context or input figure should not be used as an output allowance. Context, input and output limits describe different constraints. Do not add the figures together to infer a larger window. Check the endpoint's accounting for conversation history, system instructions, tools, media and generated tokens; a documented maximum is not a guarantee of answer quality or a measured usable capacity. ### What can Muse Spark Open 816B A42B receive and generate? Documented inputs: **text, image, audio, and video**. [Source](https://dev.meta.ai/docs/muse-spark-open) Documented outputs: **text**. [Source](https://dev.meta.ai/docs/muse-spark-open) Input support and output generation are separate capabilities. Reading an image does not establish image generation; a text output listing does not establish speech output. File sizes, formats and media limits depend on the documented implementation. ### What license is recorded for Muse Spark Open 816B A42B? The recorded license label is **Muse Model License**. [Source](https://dev.meta.ai/docs/muse-spark-open) Read the linked license and associated use terms for this exact version before redistribution, commercial deployment or fine-tuning. A label alone does not establish every permission, restriction or obligation, and it does not determine the terms of a separate hosted service. ### What distinguishes Muse Spark Open 816B A42B? These details come from the linked publisher material for this model. Reported capabilities and training descriptions are publisher statements, not independent test results. **Architecture and active parameters**: Meta describes an 816-billion-parameter mixture-of-experts model with about 42 billion active parameters per token. Total and active counts are different; neither is an independent memory benchmark. [Source](https://dev.meta.ai/docs/muse-spark-open) **Hosting route**: This open checkpoint is not a Meta Model API model. It is intended for self-hosting or access through a separate serving provider. [Source](https://dev.meta.ai/docs/models) **Hardware planning**: Meta estimates about 1.65 TB for BF16 weights and about 537 GB for NVFP4 weights. Its examples use eight GPUs for BF16 or four Blackwell GPUs for NVFP4; runtime memory and compatibility require separate planning. [Source](https://dev.meta.ai/docs/muse-spark-open) **Catalog qualifications:** Hosted through third-party partners; not assumed to be a native Meta API model. Custom license details require separate review. **Evidence still missing here:** knowledge and training-data cutoffs and tested hardware requirements. Missing documentation is not evidence that a feature is unavailable. Headless Thrones has not independently benchmarked this model for this reference. Evaluate it on representative work before choosing it; these facts do not rank its accuracy, reliability, speed or suitability against another model. Keep the exact version in view ## Other Muse Spark models from Meta. These are separate records in the same publisher family. A shared family name does not establish equal capabilities or interchangeable licenses. ### [Muse Spark 1.1](https://headlessthrones.com/models/muse-spark-1-1) API · 1,048,576 context tokens ### [Muse Spark 1.2](https://headlessthrones.com/models/muse-spark-1-2) API · 1,048,576 context tokens ### [Muse Spark 1.2 Contributor](https://headlessthrones.com/models/muse-spark-1-2-contributor) API · 1,048,576 context tokens ### [Muse Spark 1.3](https://headlessthrones.com/models/muse-spark-1-3) API · 1,048,576 context tokens ### [Muse Spark 1.3 Contributor](https://headlessthrones.com/models/muse-spark-1-3-contributor) API · 1,048,576 context tokens