What should you check first in this article?

Muse Spark Open is a multimodal model with 816B total and 42B active parameters. Meta and Apple's NVFP4 MLX path uses two M5 Ultra Mac Studios; oMLX 0.7.1rc1 reports a 370 GB oQ3e run on one 512 GB M3 Ultra Mac.

How much memory does it need?

Checkpoint size varies by precision and quantization. Meta's model card lists at least 700 GB VRAM for NVFP4 GPU deployment and 2 TB for BF16, with additional memory for KV cache and activations. These figures describe large server deployment requirements, not consumer-GPU compatibility.

When is it better not to choose this?

The quantization and inference engines differ, so do not compare the performance figures directly. [Apple's MLX setup](https://machinelearning.apple.com/research/mlx-distributed-inference) · [oMLX 0.7.1rc1](https://github.com/jundot/omlx/releases/tag/v0.7.1rc1)

Where can you check the conditions and sources?

Muse Spark Open is a multimodal language model that accepts text, images, video, and audio and returns text. It has 816B total parameters, 42B active per token, and supports up to a 1M-token context at inference. It targets coding and agent work, but local deployment starts with checking whether you have enough memory for the weights and runtime headroom. [Meta model card](https://huggingface.co/meta-models/Muse-Spark-Open-816B-A42B)

Executable programs and extensions

Can you run Muse Spark Open on one Mac?

Separate the single-Mac oQ3e path from distributed NVFP4.

Muse Spark Open is a multimodal model with 816B total and 42B active parameters. Meta and Apple's NVFP4¹ MLX² path uses two M5 Ultra Mac Studios; oMLX 0.7.1rc1 reports a 370 GB oQ3e run on one 512 GB M3 Ultra Mac.

What does this model take in and return?

Muse Spark Open is a multimodal language model that accepts text, images, video, and audio and returns text. It has 816B total parameters, 42B active per token³, and supports up to a 1M-token context at inference⁴. It targets coding and agent work, but local deployment starts with checking whether you have enough memory for the weights and runtime⁵ headroom. Meta model card

Checkpoint⁶ size varies by precision and quantization⁷. Meta's model card lists at least 700 GB VRAM⁸ for NVFP4 GPU⁹ deployment and 2 TB for BF16¹⁰, with additional memory for KV cache¹¹ and activations. These figures describe large server deployment requirements, not consumer-GPU compatibility.

A desk with a compact desktop and PC tower beside a monitor — Budget for both model weights and runtime headroom.

Choose between one 512 GB Mac and multiple M5 Ultra systems

The MLX distributed run described by Meta and Apple uses two M5 Ultra Mac Studios, each with 512 GB of unified memory¹². Separately, oMLX 0.7.1rc1 reports running a 370 GB oQ3e checkpoint on a single M3 Ultra Mac with 512 GB of unified memory.

The quantization and inference engines differ, so do not compare the performance figures directly. Apple's MLX setup · oMLX 0.7.1rc1

Distributed inference requires macOS 26.2 or later and direct Thunderbolt 5 RDMA links between the Macs. First list each Mac, its Thunderbolt IP, and RDMA device in the hostfile. Meta's mlx.distributed_config instructions explain how to create hostfile.json; the existing JACCL guide also covers connection setup and validation. Run the server command below on the primary Mac where that file is saved.

Apple MLX distributed server command

mlx.launch --hostfile hostfile.json --backend jaccl \
  --env MLX_METAL_FAST_SYNCH=1 -- \
  mlx_lm.server \
    --model meta-models/Muse-Spark-Open-816B-A42B-NVFP4 \
    --port 8080

The table shows Apple's distributed NVFP4 MLX results on M5 Ultra systems at an 8,192-token context. The single-Mac oQ3e path uses a different model format and runtime, so do not compare the figures directly; test each path with your own inputs.

Apple's M5 Ultra measurements at 8,192-token context
Mac setup	Time to first token	Decode speed
2-machine pipeline	8.4 seconds	15.74 tok/s
2-machine tensor parallelism	8.2 seconds	24.99 tok/s
4-machine tensor parallelism	4.8 seconds	36.73 tok/s

Start with oMLX on one 512 GB Mac

For one Mac, use the mlx-community/Muse-Spark-Open-816B-A42B-oQ3e variant in oMLX 0.7.1rc1, separate from Meta’s distributed NVFP4 path. The model files total 370 GB, and the release notes report loading this checkpoint on an M3 Ultra Mac with 512 GB. oMLX is a release candidate, not the stable release. oMLX release notes

In the oMLX app, open Models > Downloader and select Muse Spark Open oQ3e from the Recommended tab. Check that the disk has room for the 370 GB files before downloading, then load the model from the list. Start with a short text prompt¹³ and modest context; confirm it answers without a memory warning before adding images or longer context. Model recommendations

For this M3 Ultra 512 GB oQ3e run, oMLX 0.7.1rc1 reports 225 tok/s prefill¹⁴ and 21.3 tok/s decode¹⁵. The release notes do not include the full prompt and output conditions, so do not compare these directly with Apple’s 8,192-token M5 Ultra table. Repeat your own documents and record model loading, prefill, and decode separately to evaluate your workload.

If the model is missing or fails to load, confirm that you selected the oQ3e variant and that its download completed. If the files are present but a memory warning appears, unload other models and retry with a shorter context. Check the first load error in oMLX logs and verify that your version lists support for this model.

Muse support and measurements in oMLX 0.7.1rc1

A workspace with two compact desktops connected by cable — MLX distributes the model across Macs linked by Thunderbolt 5 RDMA.

Check modality support and license terms

Meta’s model card lists text, image, video, and audio inputs with text output. Image, video, and audio request formats vary by MLX adapter or runtime, so follow the examples in the documentation for your chosen setup.

The weights are released under the Muse Model License. Before using them in a service or redistributing the model, check the applicable terms in the full license.

A meeting table with a laptop, microphone, and notebooks — Choose the input media, then test text, images, video, or audio.

Terminology notes

NVFP4 — A 4-bit floating-point data format defined by NVIDIA. Support depends on GPU generation, model, and software implementation.
Back to the text
MLX — Apple’s machine-learning framework, which runs models using memory shared by the CPU and GPU on Apple Silicon.
Back to the text
Token — A unit into which a model divides input or output for processing.
Back to the text
Inference — The process of using a trained model to compute an output for an input. Here, local inference means running the model on the user’s device.
Back to the text
Runtime — The software environment that provides facilities needed while a program runs. In local AI, it can also refer to a model execution engine.
Back to the text
Checkpoint — A file containing saved model weights and related state. Versions or tasks in one model family may use different checkpoints.
Back to the text
Quantization — A way to represent model weights or values with fewer bits, reducing storage and memory use. The coarser representation can also affect accuracy.
Back to the text
VRAM — Memory used by a graphics card’s GPU to store model weights and intermediate values.
Back to the text
GPU — A processor designed to handle many calculations in parallel. It performs model computations during AI inference.
Back to the text
BF16 — A 16-bit floating-point format for model values. Its exponent has the same width as FP32, giving it a similarly wide value range.
Back to the text
KV cache — Memory that stores attention keys and values from earlier tokens for reuse during later token generation. Its size depends on context length and batch size.
Back to the text
Unified memory — An architecture where the CPU and GPU share one physical memory pool.
Back to the text
Prompt — The input text sent to a model. For generative models it describes the desired result and strongly shapes the output.
Back to the text
Prefill — The stage where an LLM reads the input prompt and computes representations for its tokens. Longer prompts contain more tokens to process.
Back to the text
Token decoding — The stage that generates successive output tokens using the input and the response produced so far.
Back to the text

Frequently asked questions

What should you check first in this article?: Muse Spark Open is a multimodal model with 816B total and 42B active parameters. Meta and Apple's NVFP4 MLX path uses two M5 Ultra Mac Studios; oMLX 0.7.1rc1 reports a 370 GB oQ3e run on one 512 GB M3 Ultra Mac.
How much memory does it need?: Checkpoint size varies by precision and quantization. Meta's model card lists at least 700 GB VRAM for NVFP4 GPU deployment and 2 TB for BF16, with additional memory for KV cache and activations. These figures describe large server deployment requirements, not consumer-GPU compatibility.
When is it better not to choose this?: The quantization and inference engines differ, so do not compare the performance figures directly. [Apple's MLX setup](https://machinelearning.apple.com/research/mlx-distributed-inference) · [oMLX 0.7.1rc1](https://github.com/jundot/omlx/releases/tag/v0.7.1rc1)
Where can you check the conditions and sources?: Muse Spark Open is a multimodal language model that accepts text, images, video, and audio and returns text. It has 816B total parameters, 42B active per token, and supports up to a 1M-token context at inference. It targets coding and agent work, but local deployment starts with checking whether you have enough memory for the weights and runtime headroom. [Meta model card](https://huggingface.co/meta-models/Muse-Spark-Open-816B-A42B)