← AI Hashrate 中文

Can DGX Spark (GH200 480GB) run Qwen3.5-397B-A17B?

NVIDIA · 480 GB LPDDR5X · 500 GB/s bandwidth · Alibaba · 397.0B (active 17.0B) · MoE · model ctx up to 256K

BB · GoodComposite score: VRAM headroom × 0.5 + tok/s speed × 0.5. Q4 @ 8K context.
CUDA

Yes — the DGX Spark (GH200 480GB) (480 GB LPDDR5X) can run Qwen3.5-397B-A17B. At Q4 with the default 8K context it needs ≈ 219.78 GB VRAM (weights 218.35 GB + KV cache 0.43 GB + 1 GB runtime overhead), which fits within 480 GB. Decode at Q4/8K: ≈ 18.7 tok/s (estimated, batch 1). FP16 needs ≈ 795.42 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

QuantContextVRAM neededFits?tok/s (decode)
Q44K219.56 GBYes18.7 ± est.
Q48K default219.78 GBYes18.7 ± est.
Q432K221.05 GBYes18.7 ± est.
FP164K795.21 GBNo5.4 ± est.
FP168K default795.42 GBNo5.4 ± est.
FP1632K796.7 GBNo5.3 ± 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

QuantWeightsKV cacheOverheadTotal neededDGX Spark (GH200 480GB) VRAM
Q4218.35 GB0.43 GB1 GB219.78 GB480 GB
FP16794.0 GB0.43 GB1 GB795.42 GB480 GB

Other GPUs that run Qwen3.5-397B-A17B

No catalog GPU fully fits Qwen3.5-397B-A17B at Q4/8K — see the model page for offload estimates. model page

All GPUs for Qwen3.5-397B-A17B →

Other models for the DGX Spark (GH200 480GB)

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