Meta · open-weights

Muse Spark Open

Muse Spark Open is an open-weights language model from Meta, released October 15, 2026. It has 822B parameters (mixture-of-experts, 42B active). On the Epoch Capabilities Index it scores 155.4, rank 42 of 256 rated models.

Check it against your own PC →

Lab

Can you run Muse Spark Open locally?

Even with 32 GB of VRAM and 32 GB of RAM it is too large to run at a usable speed on a single consumer GPU. Estimates assume 32 GB of DDR5 RAM, a 16K context and llama.cpp-style GGUF quantization. Enter your exact hardware in the live app.

GPU memory	Verdict	Best quant	VRAM	RAM	tok/s
6 GB VRAM	Too large	–	–	–	–
8 GB VRAM	Too large	–	–	–	–
12 GB VRAM	Too large	–	–	–	–
16 GB VRAM	Too large	–	–	–	–
24 GB VRAM	Too large	–	–	–	–
32 GB VRAM	Too large	–	–	–	–

Ready-made quantizations on Hugging Face: unsloth/Muse-Spark-Open-816B-A42B-GGUF · ggml-org/Muse-Spark-Open-816B-A42B-GGUF · Interchained/Muse-Spark-Open-816B-A42B-GGUF · Solstice-AI/Muse-Spark-Open-816B-A42B-GGUF-UltraOptimised-Curated-Protium · meta-models/Muse-Spark-Open-816B-A42B-NVFP4 · unsloth/Muse-Spark-Open-816B-A42B-NVFP4

Strengths by category

ModelTape's 0–100 fit on Epoch AI benchmark results.

Category	Score	Based on
Math	53.0	OTIS Mock AIME 2024-2025, FrontierMath-Tiers-1-3-v2-Private, FrontierMath-Tier-4-v2-Private
Knowledge	77.1	GPQA diamond, SimpleQA Verified

Benchmark results

Benchmark (Epoch AI)	Score
OTIS Mock AIME	98.1%
GPQA Diamond	88.3%
FrontierMath (Tiers 1–3)	73.5%
SimpleQA Verified	56.6%
FrontierMath-Tier-4-v2-Private	31.7%

Links

Announcement · Hugging Face: meta-models/Muse-Spark-Open-816B-A42B · Epoch AI benchmarks

Models with a similar score

DeepSeek V4.1 Flash 1017 · Gemini 3.8 Flash · Kimi K3 · Gemini 3.7 Flash · Muse Spark 1.3 · GPT-5.4 · Grok 4.6 · Qwen 3.8 Max