Can H200 141GB SXM5 run DeepSeek-R1-0528-Qwen3-8B?

NVIDIA · 141 GB HBM3e · 4800 GB/s bandwidth · DeepSeek · 8.2B · model ctx up to 128K

Yes — the H200 141GB SXM5 (141 GB HBM3e) can run DeepSeek-R1-0528-Qwen3-8B. At Q4 with the default 8K context it needs ≈ 7.15 GB VRAM (weights 4.51 GB + KV cache 1.64 GB + 1 GB runtime overhead), which fits within 141 GB. Decode at Q4/8K: ≈ 372.5 tok/s (estimated, batch 1). FP16 (≈ 19.04 GB) also fits, at ≈ 102.4 tok/s (estimated). Estimates come from memory-bandwidth math; rows tagged measured override estimates. Relative ranking is more reliable than absolute tok/s. Methodology.

Fit & speed by quant and context

Quant	Context	VRAM needed	Fits?	tok/s (decode)
Q4	4K	6.33 GB	Yes	372.5 est.
Q4	8K default	7.15 GB	Yes	372.5 est.
Q4	32K	12.07 GB	Yes	372.5 est.
FP16	4K	18.22 GB	Yes	102.4 est.
FP16	8K default	19.04 GB	Yes	102.4 est.
FP16	32K	23.96 GB	Yes	102.4 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	H200 141GB SXM5 VRAM
Q4	4.51 GB	1.64 GB	1 GB	7.15 GB	141 GB
FP16	16.4 GB	1.64 GB	1 GB	19.04 GB	141 GB

Other GPUs that run DeepSeek-R1-0528-Qwen3-8B

• B200 192GB SXM — ≈620.8 tok/s Q4 est.
• MI300X 192GB — ≈411.3 tok/s Q4 est.
• H100 80GB SXM5 — ≈260.0 tok/s Q4 est.
• Gaudi 3 128GB — ≈254.2 tok/s Q4 est.
• A100 80GB SXM4 — ≈158.2 tok/s Q4 est.
• RTX 5090 32GB — ≈139.1 tok/s Q4 est.
• RTX 4090 24GB — ≈78.2 tok/s Q4 est.
• RTX 4090D 24GB (China) — ≈78.2 tok/s Q4 est.

All GPUs for DeepSeek-R1-0528-Qwen3-8B →

Other models for the H200 141GB SXM5

• Qwen3-0.6B — ≈3818.2 tok/s Q4 est.
• gemma-3-1b-it — ≈3054.5 tok/s Q4 est.
• Llama-3.2-1B-Instruct — ≈2545.5 tok/s Q4 est.
• Qwen3-1.7B — ≈1527.3 tok/s Q4 est.
• Qwen3-30B-A3B — ≈1018.2 tok/s Q4 est.
• Qwen3-Coder-30B-A3B-Instruct — ≈1018.2 tok/s Q4 est.
• GLM-4.7-Flash — ≈1018.2 tok/s Q4 est.
• Qwen3-30B-A3B-Instruct-2507 — ≈1018.2 tok/s Q4 est.

All models on the H200 141GB SXM5 →