Can H100 80GB SXM5 run Mistral-Medium-3.5-128B?

NVIDIA · 80 GB HBM3 · 3350 GB/s bandwidth · Mistral · 128.0B · model ctx up to 256K

No — Mistral-Medium-3.5-128B at Q4 needs ≈ 84.2 GB even at 4K context, beyond the H100 80GB SXM5's 80 GB HBM3. It would only run with heavy CPU/disk offload. Decode at Q4/8K: ≈ 32.4 tok/s (estimated, batch 1). FP16 needs ≈ 282.6 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.

Fit & speed by quant and context

Quant	Context	VRAM needed	Fits?	tok/s (decode)
Q4	4K	84.2 GB	No	43.0 est.
Q4	8K default	97.0 GB	No	32.4 est.
Q4	32K	173.8 GB	No	10.1 est.
FP16	4K	269.8 GB	No	1.2 est.
FP16	8K default	282.6 GB	No	1.0 est.
FP16	32K	359.4 GB	No	—

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	H100 80GB SXM5 VRAM
Q4	70.4 GB	25.6 GB	1 GB	97.0 GB	80 GB
FP16	256.0 GB	25.6 GB	1 GB	282.6 GB	80 GB

Other GPUs that run Mistral-Medium-3.5-128B

• B200 192GB SXM — ≈39.8 tok/s Q4 est.
• MI300X 192GB — ≈26.3 tok/s Q4 est.
• H200 141GB SXM5 — ≈23.9 tok/s Q4 est.
• Gaudi 3 128GB — ≈16.3 tok/s Q4 est.
• Mac Studio M2 Ultra 192GB — ≈4.0 tok/s Q4 est.
• MacBook Pro M4 Max 128GB — ≈2.7 tok/s Q4 est.
• MacBook Pro M3 Max 128GB — ≈2.0 tok/s Q4 est.

All GPUs for Mistral-Medium-3.5-128B →

Other models for the H100 80GB SXM5

• MiniCPM5-1B — ≈1973.9 tok/s Q4 est.
• Qwen3.5-2B — ≈1065.9 tok/s Q4 est.
• GLM-4.7-Flash — ≈710.6 tok/s Q4 est.
• Qwen3-30B-A3B-Instruct-2507 — ≈710.6 tok/s Q4 est.
• Qwen3-Next-80B-A3B-Instruct — ≈710.6 tok/s Q4 est.
• Qwen3-Coder-Next — ≈710.6 tok/s Q4 est.
• Qwen3.5-35B-A3B — ≈710.6 tok/s Q4 est.
• Qwen3.6-35B-A3B — ≈710.6 tok/s Q4 est.

All models on the H100 80GB SXM5 →