← AI Hashrate 中文

Can B200 192GB SXM run Hy3?

NVIDIA · 192 GB HBM3e · 8000 GB/s bandwidth · Tencent · 295.0B (active 21.0B) · MoE · model ctx up to 256K

✅ Yes — B200 192GB SXM (192 GB) can run Hy3 at Q4 (8K)
VRAM: 167.45 GB needed of 192 GB
Speed: ~242.4 tok/s (estimated)
BB · GoodComposite score: VRAM headroom × 0.5 + tok/s speed × 0.5. Q4 @ 8K context.
CUDA
Estimates from memory-bandwidth formula; rows marked "measured" override. Methodology.

Fit & speed by quant and context

QuantContextVRAM neededFits?tok/s (decode)
Q44K165.35 GBYes242.4 ± est.
Q48K default167.45 GBYes242.4 ± est.
Q432K180.05 GBYes242.4 ± est.
FP164K593.1 GBNo20.0 ± est.
FP168K default595.2 GBNo19.8 ± est.
FP1632K607.8 GBNo19.0 ± 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 neededB200 192GB SXM VRAM
Q4162.25 GB4.2 GB1 GB167.45 GB192 GB
FP16590.0 GB4.2 GB1 GB595.2 GB192 GB

Other GPUs that run Hy3

All GPUs for Hy3 →

Other models for the B200 192GB SXM

All models on the B200 192GB SXM →