Muse Spark Open 816B-A42B
816B-A42BMeta Muse
Meta's largest open-weight model, released in October under the Muse Model License. It is an 816B mixture-of-experts model with about 42B parameters active per token, built for coding and agent work, with image and audio input. It is a single-machine model only at its smallest build. At Q2_K (ggml-org's 304 GB file) it needs about 319 GB at 4K context and still only 327 GB at 128K, because just 16 of its 62 layers keep a growing cache — comfortable on a 512 GB Mac Studio and tight across four 80 GB H100s. The full 1M window brings it to about 389 GB. Q4_K_S (470 GB) and Q8_0 (872 GB) fit no single machine in this index. The vision and audio encoders ship as separate files of 2–4 GB and are not included. Sizes come from the published GGUF files; Meta's benchmark claims are its own and no per-level quality loss has been published. The licence is not a standard open-source one: Meta's repository is gated, so downloads need an accepted access request; products built on it must show "Built with Muse Spark Open"; and a company with over $20M in annual revenue must sign a separate agreement with Meta before offering it as an API service.
1049K
Max Context
3
Quant Variants
GGUF Q8_0
Best Quality
—
Accuracy Retained
Quantization Variants
Per-quant VRAM, quality loss, and inference speed on RTX 4090
Measured = run on this site · Estimated = not run here: calculated, or a figure this site has not verified · Community = public reports
Similar models
Compare with Muse GlimmerMuse Glimmer 30B
Meta Muse
Meta's first open-weight model since Llama 4, released in August under Apache 2.0 and built to run on one card. It is a dense model: a 28B text decoder that reads images through a separate vision encoder and answers in text. Three of every four layers use a 2K sliding window and the rest share just 2 KV heads, so long context is unusually cheap. At Q4_K_M it needs about 17.7 GB at 4K context and only 19.4 GB at its full 128K window, comfortable on a 24 GB card either way. Q5_K_M is tight on 24 GB, and Q8_0 wants a 32 GB card. The vision encoder (about 1.4 GB as a GGUF mmproj file) and Meta's optional speculative-decoding drafter load on top and are not included. Sizes come from the published GGUF file sizes; the Q4_K_M row is Meta's own official build. Meta's benchmark claims are its own, and no per-level quality loss has been published.
Nemotron 3.5 Bolt 74B-A9B
NVIDIA Nemotron
NVIDIA's Nemotron 3.5 Bolt is a 74B mixture-of-experts model with about 9B parameters active per token, built to fit a 24 GB card. It is compressed from Nemotron 3.5 Super VL and ships in one build, NVIDIA's own native 2-bit file of 22.0 GB. Only 8 of its 88 layers keep a growing cache, so context costs little: about 23.1 GB at 4K, 24.1 GB at 128K and 25.2 GB at its full 256K window. That is the 24 GB NVIDIA designed for, but over this site's comfortable line, so a 24 GB card reads as tight; a 32 GB card or a 48 GB Mac runs it comfortably. The image encoder is a separate 1.7 GB file. NVIDIA's benchmark claims are its own, and no quality loss for the 2-bit build has been published.
Nemotron 3.5 Super VL 120B-A12B
NVIDIA Nemotron
NVIDIA's Nemotron 3.5 Super VL is a 120B mixture-of-experts vision-language model with about 12B parameters active per token. It is built for coding, tool use and document work, with a context window of up to 1M tokens. Only 8 of its 88 layers keep a growing cache, so even the full window is cheap: at Q4_0 (ggml-org's 71.8 GB file) it needs about 75 GB at 4K and 84 GB at 1M. That runs comfortably on a 128 GB Mac or a DGX Spark 128G, and tightly on one 80 GB card. Q8_0 (128 GB) wants a 256 GB Mac. The vision encoder is a separate 1–1.7 GB file. Sizes come from the published GGUF files; NVIDIA's benchmark claims are its own and no per-level quality loss has been published.
Mistral Large 3 675B Instruct
Mistral AI
Mistral 3 flagship MoE (41B active / 675B total) with vision encoder. FP8 on 8×H200; GGUF quant for research clusters only. Its attention is DeepSeek-style MLA, so the real KV cache is smaller than the figure below, which uses an unverified conventional layout. At 32K that is a rounding error beside ~390 GB of weights; at the full 256K window, treat the cache figure as an upper bound.
Running Muse Spark Open 816B-A42B locally
At Q4_K_S and 4K of context, Muse Spark Open 816B-A42B needs about 492.4 GB — 446.9 GB of weights, 0.70 GB of KV cache and a 44.8 GB activation buffer. No card in this index clears that comfortably at Q4_K_S, so at that level it is a multi-GPU or CPU-offload proposition. Its smaller Q2_K build needs about 318.9 GB, which the Mac M5 Ultra 512G clears comfortably. These are calculated figures, not measurements: the estimate stops counting a card as comfortable at 88% of its VRAM, which is roughly the room a desktop session needs.
What longer context costs
Going from 4K to 32K adds about 1.75 GB, taking the total to 494.3 GB. That is gentler than the parameter count suggests: only 16 of 62 layers keep a KV cache that grows with context, and 46 more cap theirs at a sliding window, so the cache scales at a fraction of the usual rate. The model's native window is 1M; holding all of it at Q4_K_S would need about 563 GB.
Which build to download
This index only tracks Muse Spark Open 816B-A42B in GGUF, across 3 levels (Q2_K, Q4_K_S, Q8_0). No per-level perplexity sweep has been published for this model, so the quality column is empty rather than estimated — within one model, more bits per weight is the only ordering the data supports. No throughput has been measured for this model on this site's hardware. GGUF runs on llama.cpp and Ollama across NVIDIA, AMD and Apple silicon; AWQ and GPTQ target vLLM on CUDA and ROCm; EXL2 is ExLlamaV2 — CUDA only, and now archived, so local use rather than serving.
Common questions
- How much VRAM does Muse Spark Open 816B-A42B need?
- About 492.4 GB at Q4_K_S with 4K of context and batch 1, rising to roughly 494.3 GB at 32K. That figure is weights plus KV cache plus a 10% activation buffer, calculated from the model's architecture rather than measured on a card.
- Can Muse Spark Open 816B-A42B run on a single GPU?
- Not comfortably on any single card in this index at Q4_K_S. It needs about 492.4 GB, which means splitting across GPUs or offloading layers to system RAM. Its smaller Q2_K build needs about 318.9 GB, which the Mac M5 Ultra 512G clears comfortably.
- Which quantization of Muse Spark Open 816B-A42B should I use?
- No published quality comparison exists for this model, so pick by footprint: Q2_K is the largest level here that fits a single machine in this index, and a higher bits-per-weight level is more faithful. All 3 levels in the index are Q2_K, Q4_K_S, Q8_0. Note this is a mixture-of-experts model — all parameters must be resident even though only a fraction are active per token, so the memory cost follows the total, not the active count.
Where this model fits
Sized at Q4_K_S with a 4K context window, smallest card first. Comfortable means the estimate uses at most 88% of the memory.
- Ships in
- GGUF
- Try it yourself
- Size it in the calculatorGenerate the run command
How to actually run this
Deployment guides for this model and this class of hardware.
This page's figures change when the model or the runtime does.Last updated 2026-10-18 RSS → /feed.xml