← AI Hashrate

Can Gaudi 3 128GB run Qwen3.6-27B?

Intel · 128 GB HBM2e · 3276 GB/s bandwidth · Alibaba · 27.8B · model ctx up to 256K

Yes — the Gaudi 3 128GB (128 GB HBM2e) can run Qwen3.6-27B. At Q4 with the default 8K context it needs ≈ 21.85 GB VRAM (weights 15.29 GB + KV cache 5.56 GB + 1 GB runtime overhead), which fits within 128 GB. Decode at Q4/8K: ≈ 75.0 tok/s (estimated, batch 1). FP16 (≈ 62.16 GB) also fits, at ≈ 20.6 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

QuantContextVRAM neededFits?tok/s (decode)
Q44K19.07 GBYes75.0 est.
Q48K default21.85 GBYes75.0 est.
Q432K38.53 GBYes75.0 est.
FP164K59.38 GBYes20.6 est.
FP168K default62.16 GBYes20.6 est.
FP1632K78.84 GBYes20.6 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 neededGaudi 3 128GB VRAM
Q415.29 GB5.56 GB1 GB21.85 GB128 GB
FP1655.6 GB5.56 GB1 GB62.16 GB128 GB

Other GPUs that run Qwen3.6-27B

All GPUs for Qwen3.6-27B →

Other models for the Gaudi 3 128GB

All models on the Gaudi 3 128GB →