Can DGX Spark (GH200 480GB) run MiMo-V2.5?

NVIDIA · 480 GB LPDDR5X · 500 GB/s bandwidth · Xiaomi · 310.0B (active 15.0B) · MoE · model ctx up to 1024K

CUDA

Yes — the DGX Spark (GH200 480GB) (480 GB LPDDR5X) can run MiMo-V2.5. At Q4 with the default 8K context it needs ≈ 171.88 GB VRAM (weights 170.5 GB + KV cache 0.38 GB + 1 GB runtime overhead), which fits within 480 GB. Decode at Q4/8K: ≈ 21.2 tok/s (estimated, batch 1). FP16 needs ≈ 621.38 GB — does not fit on this card. Estimates come from memory-bandwidth math; rows tagged measured override estimates. Relative ranking is more reliable than absolute tok/s. Methodology. Buy DGX Spark (GH200 480GB) on Amazon →

Fit & speed by quant and context

Quant	Context	VRAM needed	Fits?	tok/s (decode)
Q4	4K	171.69 GB	Yes	21.2 ± est.
Q4	8K default	171.88 GB	Yes	21.2 ± est.
Q4	32K	173.0 GB	Yes	21.2 ± est.
FP16	4K	621.19 GB	No	10.0 ± est.
FP16	8K default	621.38 GB	No	9.9 ± est.
FP16	32K	622.5 GB	No	9.9 ± est.

Fits = weights + KV(ctx) + 1 GB overhead ≤ 95% of VRAM. Measured anchors are context-agnostic; the fit verdict is recomputed per context. A missing tok/s means the model is far beyond this card (offload-only territory).

VRAM breakdown at 8K context

Quant	Weights	KV cache	Overhead	Total needed	DGX Spark (GH200 480GB) VRAM
Q4	170.5 GB	0.38 GB	1 GB	171.88 GB	480 GB
FP16	620.0 GB	0.38 GB	1 GB	621.38 GB	480 GB

Other GPUs that run MiMo-V2.5

• B200 192GB SXM — ≈339.4 tok/s Q4 est.
• MI300X 192GB — ≈224.8 tok/s Q4 est.
• Mac Studio M2 Ultra 192GB — ≈33.9 tok/s Q4 est.

All GPUs for MiMo-V2.5 →

Other models for the DGX Spark (GH200 480GB)

• MiniCPM5-1B — ≈294.6 tok/s Q4 est.
• Qwen3.5-2B — ≈159.1 tok/s Q4 est.
• DeepSeek-R1-0528-Qwen3-8B-layer-mix-bpw-3.8-mlx — ≈138.3 tok/s Q4 est.
• Qwen3.6-35B-A3B — ≈106.1 tok/s Q4 est.
• Qwen3.5-35B-A3B — ≈106.1 tok/s Q4 est.
• GLM-4.7-Flash — ≈106.1 tok/s Q4 est.
• Qwen3-30B-A3B-Instruct-2507 — ≈106.1 tok/s Q4 est.
• Nemotron-3-Nano-Omni-30B-A3B-Reasoning-BF16 — ≈106.1 tok/s Q4 est.

All models on the DGX Spark (GH200 480GB) →