← AI Hashrate

Can RTX 5090 32GB run gemma-3-27b-it?

NVIDIA · 32 GB GDDR7 · 1792 GB/s bandwidth · Google · 27.4B · model ctx up to 128K

Yes — the RTX 5090 32GB (32 GB GDDR7) can run gemma-3-27b-it. At Q4 with the default 8K context it needs ≈ 21.51 GB VRAM (weights 15.07 GB + KV cache 5.44 GB + 1 GB runtime overhead), which fits within 32 GB. Decode at Q4/8K: ≈ 41.9 tok/s (estimated, batch 1). FP16 needs ≈ 61.24 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

QuantContextVRAM neededFits?tok/s (decode)
Q44K18.79 GBYes41.9 est.
Q48K default21.51 GBYes41.9 est.
Q432K37.83 GBNo85.7 est.
FP164K58.52 GBNo9.8 est.
FP168K default61.24 GBNo9.0 est.
FP1632K77.56 GBNo5.6 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 neededRTX 5090 32GB VRAM
Q415.07 GB5.44 GB1 GB21.51 GB32 GB
FP1654.8 GB5.44 GB1 GB61.24 GB32 GB

Other GPUs that run gemma-3-27b-it

All GPUs for gemma-3-27b-it →

Other models for the RTX 5090 32GB

All models on the RTX 5090 32GB →