Meta, United States · Released October 15, 2026Muse Spark Open at a glance
- Made by
- Meta, United States
- Released
- October 15, 2026
- Parameters
- 822.0B, 42B active
- Context window
- 1M tokens
Muse Spark Open VRAM Requirements
Meta, United States · Released October 15, 2026New
Muse Spark Open has 822.0B parameters (42B active per token). In BF16 its weights alone take 1531 GiB; quantized to 4-bit, about 461 GiB. Each 32K-token request adds 2.36 GiB of KV cache. Cheapest way to run it today: 5 × RTX PRO 6000 Blackwell (96 GB) at $6.67 per hour.
Size your own setup in the calculator →VRAM by precision and context length
Weights plus the KV cache of one request (BF16 cache), in GiB. Add 1 to 3 GiB for the inference engine itself.
| Weights format | Weights only | + 8K context | + 32K context | + 128K context | + 1M context |
|---|---|---|---|---|---|
| NVFP4 experts + BF16 (native) | 496 | 497 | 499 | 505 | 561 |
| BF16 / FP16 | 1531 | 1532 | 1534 | 1540 | 1596 |
| FP8 / INT8 | 769 | 770 | 771 | 777 | 833 |
| 4-bit (GGUF Q4_K_M) | 461 | 462 | 464 | 470 | 526 |
How many GPUs to run Muse Spark Open
Fewest GPUs that fit the weights plus one 8K-token request. Data-center GPUs use vLLM (90% of memory usable), the others llama.cpp. Several GPUs split the model with tensor or pipeline parallelism.
| GPU | NVFP4 experts + BF16 (native) | BF16 / FP16 | 8-bit | 4-bit |
|---|---|---|---|---|
| GeForce RTX 4090 24 GB | Not on this GPU | 96 (several machines) | 48 (several machines) | 24 (several machines) |
| GeForce RTX 5090 32 GB | Not on this GPU | 64 (several machines) | 32 (several machines) | 16 (several machines) |
| RTX PRO 6000 Blackwell 96 GB | Not on this GPU | 24 (several machines) | 16 (several machines) | 5 |
| Mac M5 Max 128 GB 96 GB usable | Not on this GPU | 24 (several machines) | 16 (several machines) | 5 |
| Ryzen AI Max+ 395 128 GB 96 GB usable | Not on this GPU | 24 (several machines) | 16 (several machines) | 5 |
| DGX Spark (GB10) 128 GB | Not on this GPU | 16 (several machines) | 7 | 4 |
| NVIDIA H100 80 GB | Not on this GPU | 24 (several servers) | 16 (several servers) | 7 |
| NVIDIA H200 141 GB | Not on this GPU | 16 (several servers) | 7 | 4 |
| NVIDIA B200 (HGX) 180 GB | 4 | 16 (several servers) | 6 | 3 |
8-bit: FP8 with vLLM (INT8 on GPUs without FP8), Q8_0 with llama.cpp. 4-bit: AWQ / GPTQ with vLLM, Q4_K_M with llama.cpp, both counted at the Q4_K_M size. Not on this GPU: the official NVFP4 experts + BF16 weights do not run there (llama.cpp needs a GGUF version, and vLLM needs a GPU with FP4 support).
Cheapest way to run Muse Spark Open today
For each GPU: the fewest cards that fit Muse Spark Open with one 8K-token request, the most faithful weight format at that count, and the cheapest on-demand price (2026-10-16).
| Setup | Per hour | Weights | Provider | Per month | Rent |
|---|---|---|---|---|---|
| 5 × RTX PRO 6000 Blackwell 96 GB | $6.67 | Q4_K_M | Vast.ai | $4,869.10 | Rent 5 × RTX PRO 6000 Blackwell on Vast.ai (opens in a new tab) → |
| 7 × NVIDIA A100 80GB | $7.01 | 4-bit | Vast.ai | $5,115.11 | Rent 7 × NVIDIA A100 80GB on Vast.ai (opens in a new tab) → |
| 3 × AMD MI300X 192 GB | $7.17 | 4-bit | RunPod | $5,234.10 | Rent 3 × AMD MI300X on RunPod (opens in a new tab) → |
| 2 × AMD MI355X / MI350X 288 GB | $10.98 | NVFP4 experts + BF16 (official) | RunPod | $8,015.40 | Rent 2 × AMD MI355X / MI350X on RunPod (opens in a new tab) → |
| 4 × NVIDIA H200 141 GB | $15.79 | 4-bit | Vast.ai | $11,528.16 | Rent 4 × NVIDIA H200 on Vast.ai (opens in a new tab) → |
| 6 × NVIDIA H100 NVL 94 GB | $16.01 | 4-bit | Vast.ai | $11,685.84 | Rent 6 × NVIDIA H100 NVL on Vast.ai (opens in a new tab) → |
| 7 × NVIDIA H100 80 GB | $16.80 | 4-bit | Vast.ai | $12,264.00 | Rent 7 × NVIDIA H100 on Vast.ai (opens in a new tab) → |
| 2 × NVIDIA B300 / GB300 288 GB | $16.81 | 4-bit | Scaleway | $12,272.76 | Rent 2 × NVIDIA B300 / GB300 on Scaleway (opens in a new tab) → |
On-demand prices from AWS, Google Cloud, Azure, RunPod, Vast.ai, Oracle Cloud, OVHcloud, Scaleway, Verda, Akash and UpCloud, checked every hour. Serving many users needs more KV cache, so more memory: size it in the calculator. Some links are affiliate links: StudioTV may earn a commission, at no extra cost to you.
AWS, Google Cloud and UpCloud prices: Spare Cores (opens in a new tab), CC BY-SA 4.0 (opens in a new tab), shared under the same license.
KV cache: how memory grows with context
Architecture: MoE 256 experts · 16 full-attention layers + 46 sliding-window layers (2048 tokens), GQA 16 KV × 64. Its KV cache is 3.3× smaller than if every layer used full attention, because most layers only keep a short window of recent tokens. Only 42B parameters are active per token, but all 822B must sit in memory: VRAM depends on the total, speed on the active part.
| Context per request | BF16 cache | FP8 cache |
|---|---|---|
| 8K tokens | 0.86 GiB | 0.43 GiB |
| 32K tokens | 2.36 GiB | 1.18 GiB |
| 128K tokens | 8.36 GiB | 4.18 GiB |
| 1M tokens | 64.4 GiB | 32.2 GiB |
Muse Spark Open VRAM FAQ
How much VRAM does Muse Spark Open need?
In BF16 the weights alone take 1531 GiB (1644 GB). In FP8 that is 769 GiB, and about 461 GiB with 4-bit quantization (Q4_K_M); the official NVFP4 experts + BF16 checkpoint is 496 GiB. Each request then adds KV cache: 0.86 GiB at 8K tokens and 2.36 GiB at 32K (BF16 cache).
Can Muse Spark Open run on a single RTX 4090 (24 GB)?
No. Even in 4-bit it needs 24 RTX 4090s, spread over several machines. In 4-bit it needs 7 H100 GPUs.
Can Muse Spark Open run on an RTX 5090 (32 GB)?
No. Even in 4-bit it needs 16 RTX 5090s, spread over several machines.
Can Muse Spark Open run on a Mac?
No single Mac holds it, even in 4-bit: it needs 462 GiB with an 8K-token context, and the largest Mac (512 GB) gives its GPU about 384 GB.
Can Muse Spark Open run on a DGX Spark or a Ryzen AI Max+ 395 (128 GB)?
No. Even in 4-bit it needs 4 DGX Spark units or 5 Ryzen AI Max+ 395 machines (128 GB), linked over the network.
What is the cheapest way to run Muse Spark Open?
On 2026-10-16, the cheapest on-demand setup is 5 × RTX PRO 6000 Blackwell (96 GB) with Q4_K_M weights on Vast.ai, at $6.67 per hour (about $4,869.10 per month). Next: 7 × NVIDIA A100 80GB with 4-bit weights on Vast.ai, at $7.01 per hour. Sized for one 8K-token request; prices are checked every hour.
How many H100 GPUs does Muse Spark Open need?
With one request and an 8K-token context: 24 in BF16, 16 in FP8 and 7 in 4-bit (vLLM, 90% of memory usable). Above 8 GPUs, that means several servers. Serving many users at once needs more memory for their KV caches: the calculator sizes that for you.
Muse Spark Open VRAM badge
For a model card or a README: the 4-bit size and the smallest GPU it fits on, linked to this page. Data as JSON: /api/models/muse-spark-open.json.
[](https://studiotvai.com/vram-requirements/muse-spark-open)Other models
Same family (large moe): MiniMax M2.7 · DeepSeek R1 · Kimi K2.6 · GLM-5.3 · GLM-5.3-Flash · DeepSeek V4 Flash · DeepSeek V4 Pro · DeepSeek V4.1 Flash · MiMo-V2.6 Pro · MiMo-V2.6 Flash · Kimi K3 · Mistral Large 4 · DeepSeek V4 Flash Vision Exp · Qwen3.8 2.4T-A95B · Ornith 1.5 397B · Atria Dawn Preview · Intern S2 397B · DeepSeek V3 · DeepSeek V3.1 · Kimi K2.7 Code · Kimi K2.5 · Kimi K2 · GLM-5.2 · GLM-5.1 · GLM-5 · MiniMax M2.5 · MiniMax M2 · MiMo-V2.5 · Ornith-1.0 397B · Qwen3 235B-A22B · Qwen3.5 397B-A17B · GLM-4.7 · GLM-4.5 · Ling-3.1-flash · Step 5 Preview
All 118 other models
Llama 3.1 8B · Mistral Small 3.2 24B · Qwen3 32B · Llama 3.3 70B · Llama 3.1 405B · Qwen3.5 4B · Qwen3.5 9B · Qwen3.8 27B · Qwen3.6 35B-A3B · Qwen3.5 122B-A10B · Qwen3.8-Flash-Next · LFM2.5 8B-A1B · LFM2 24B-A2B · Gemma 4 12B · Gemma 4 31B · Gemma 4 26B-A4B · gpt-oss 20B · gpt-oss 120B · Llama 4 Scout 17B-16E · Llama 4 Maverick 17B-128E · Kolibri-1 · Step 3.5 Flash · Muse Glimmer 30B · Nemotron 3 Nano 30B-A3B · Nemotron 3 Super 120B-A12B · MiniMax M2.7 · DeepSeek R1 · Kimi K2.6 · GLM-5.3 · GLM-5.3-Flash · DeepSeek V4 Flash · DeepSeek V4 Pro · DeepSeek V4.1 Flash · MiMo-V2.6 Pro · MiMo-V2.6 Flash · Kimi K3 · Mistral Large 4 · Qwen3-Coder-Next 80B-A3B · Qwen3-Coder 30B-A3B · Qwen3-Coder 480B-A35B · DeepSeek V4 Flash Vision Exp · Qwen3.8 2.4T-A95B · Ornith 1.5 397B · Ornith 1.5 35B-A3B · Ornith 1.5 9B · Atria Dawn Preview · Intern S2 397B · Clef · Clef Flash · D1 3B · Humanizer · JEV 27B VL · Mellum2.1 12B-A2.5B · LightOnOCR-3 4B · Spark-X2.5 4B · Agnes-3.0-Qwen · Youtu-Parsing-Omni · Agnes-3.0-Flash · Mistral 7B v0.3 · Mistral Small 3.1 24B · Mistral Nemo · Ministral 8B · DeepSeek V3 · DeepSeek R1 Qwen3 8B · DeepSeek R1 Distill Qwen 32B · DeepSeek V3.1 · DeepSeek R1 Distill Qwen 14B · DeepSeek R1 Distill Qwen 7B · DeepSeek R1 Distill Llama 8B · Qwen3 4B · Qwen3 8B · Qwen2.5 7B · Qwen3 14B · Qwen2.5 3B · Qwen3.6 27B · Qwen3.5 27B · Qwen2.5-Coder 7B · Qwen2.5 32B · Qwen2.5-Coder 14B · Qwen3.5 35B-A3B · Qwen3 30B-A3B · Qwen2.5 14B · Qwen2.5-Coder 32B · Kimi K2.7 Code · Kimi K2.5 · Kimi K2 · GLM-5.2 · GLM-5.1 · GLM-5 · MiniMax M2.5 · MiniMax M2 · MiMo-V2.5 · Ornith-1.0 397B · Qwen2.5-Coder 3B · Qwen3 235B-A22B · Qwen3.5 397B-A17B · Gemma 4 E4B · Gemma 4 E2B · Nemotron 3 Nano 4B · Nemotron 3.5 Lightning 30B-A3B · MiniCPM-V-4.7 35B-A3B · LFM2.5 24B-A2B · GLM-4.7 · Qwen3-VL 8B · Qwen2.5-VL 7B · Qwen3-VL 4B · Qwen2.5-VL 3B · Qwen2.5-VL 32B · Qwen2-VL 7B · GLM-4.5-Air · GLM-4.5 · phi-4 · Phi-3.5-mini · Phi-4-mini · Phi-3-mini-128k · Unlimited-OCR · Ling-3.1-flash · Step 5 Preview
Estimates, not guarantees: computed from the official config.json with the same engine as the LLM VRAM Calculator. Real usage depends on your engine version and settings.