← AI Hashrate 中文

Can DGX Spark (GH200 480GB) run GLM-5.2?

NVIDIA · 480 GB LPDDR5X · 500 GB/s bandwidth · Z.ai · 753.3B (active 40B) · MoE · model ctx up to 1000K

✅ Yes — DGX Spark (GH200 480GB) (480 GB) can run GLM-5.2 at Q4 (8K)
VRAM: 423.31 GB needed of 480 GB
Speed: ~8.0 tok/s (estimated)
CC · WorkableComposite score: VRAM headroom × 0.5 + tok/s speed × 0.5. Q4 @ 8K context.
CUDA
Buy DGX Spark (GH200 480GB) on Amazon →
Estimates from memory-bandwidth formula; rows marked "measured" override. Methodology.

Fit & speed by quant and context

QuantContextVRAM neededFits?tok/s (decode)
Q44K419.31 GBYes8.0 ± est.
Q48K default423.31 GBYes8.0 ± est.
Q432K447.31 GBYes8.0 ± est.
FP164K1511.6 GBNo0.6 ± est.
FP168K default1515.6 GBNo0.6 ± est.
FP1632K1539.6 GBNo0.6 ± 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
Q4414.31 GB8.0 GB1 GB423.31 GB480 GB
FP161506.6 GB8.0 GB1 GB1515.6 GB480 GB

Other GPUs that run GLM-5.2

No catalog GPU fully fits GLM-5.2 at Q4/8K — see the model page for offload estimates. model page

All GPUs for GLM-5.2 →

Other models for the DGX Spark (GH200 480GB)

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