RunsOnMy ModelsGPUs & MacsBy VRAMAPI pricesVRAM calculatorNew

Can Qwen3.6 27B run on the 2x RTX 5090?

Yes. At 8K context the best fit is F16, using about 51.5 GB of the 60.8 GB usable on the 2x RTX 5090. Estimated speed ceiling: ~33 tokens/sec.

Memory by quant and context

Quant4K ctx8K ctx32K ctx128K ctxMax tok/s (est., 8K)
2-bit (Q2_K)10.8 GB Fits11.1 GB Fits12.6 GB Fits18.6 GB Fits172
3-bit (Q3_K_M) •13.8 GB Fits14.0 GB Fits15.5 GB Fits21.5 GB Fits132
4-bit (Q4_K_M) •16.8 GB Fits17.0 GB Fits18.5 GB Fits24.5 GB Fits107
5-bit (Q5_K_M) •19.3 GB Fits19.5 GB Fits21.0 GB Fits27.0 GB Fits92
6-bit (Q6_K) •22.1 GB Fits22.4 GB Fits23.9 GB Fits29.9 GB Fits80
8-bit (Q8_0) •27.8 GB Fits28.0 GB Fits29.5 GB Fits35.5 GB Fits63
16-bit (F16/BF16) •51.2 GB Fits51.5 GB Fits53.0 GB Fits59.0 GB Tight33

• real GGUF file size from unsloth/Qwen3.6-27B-GGUF. Speed is a bandwidth ceiling (1792 GB/s), not a benchmark. Methodology.

More

Qwen3.6 27B on every GPU and Mac · Every model that runs on the 2x RTX 5090 · Try other settings in the calculator

Data: Hugging Face Hub (Qwen/Qwen3.6-27B), NVIDIA specs. Data as of 2026-11-17.