← CanIRunLocal · All hardware · All models
Meta · 816B parameters (42B active) · chat, reasoning, coding, vision.
Ollama lists muse-spark-open:cloud as a cloud model, so there is no local download. The fits below use verified Hugging Face weights.
Yes on 1 machine in this list. The smallest is 4x Mac Studio M5 Ultra (80c GPU) 256GB: Q4_K_M, about 535.5 GB, roughly 39 tokens per second, via MLX. exo # on every Mac (brew install --cask exo); load mlx-community/Muse-Spark-Open-816B-A42B-4bit with Tensor sharding + RDMA
exo # on every Mac (brew install --cask exo); load mlx-community/Muse-Spark-Open-816B-A42B-4bit with Tensor sharding + RDMA| Hardware | Runtime | Quant | Memory | Fit | Speed | Command |
|---|---|---|---|---|---|---|
| 4x Mac Studio M5 Ultra (80c GPU) 256GB | MLX | MLX 4bit | 535.5 GB | fits in fast memory | ~39 tok/s | exo # on every Mac (brew install --cask exo); load mlx-community/Muse-Spark-Open-816B-A42B-4bit with Tensor sharding + RDMA |
Estimates, not benchmarks. Model sizes are ollama.com download sizes; memory needed adds headroom for ~8k tokens of context; speeds are rough figures from memory bandwidth. Data updated 2026-10-19.
A project by OnDevice Labs · llms.txt