RunsOnMy ModelsGPUs & MacsBy VRAMAPI pricesVRAM calculatorNew

Can Qwen2.5 Coder 14B Instruct run on the RTX 3090?

Yes. At 8K context the best fit is Q8_0, using about 16.8 GB of the 22.8 GB usable on the RTX 3090. Estimated speed ceiling: ~60 tokens/sec.

Memory by quant and context

Quant4K ctx8K ctx32K ctxMax tok/s (est., 8K)
2-bit (Q2_K)6.6 GB Fits7.3 GB Fits11.8 GB Fits169
3-bit (Q3_K_M)8.2 GB Fits8.9 GB Fits13.4 GB Fits130
4-bit (Q4_K_M)9.8 GB Fits10.6 GB Fits15.1 GB Fits104
5-bit (Q5_K_M)11.2 GB Fits12.0 GB Fits16.5 GB Fits89
6-bit (Q6_K)12.7 GB Fits13.5 GB Fits18.0 GB Fits77
8-bit (Q8_0)16.1 GB Fits16.8 GB Fits21.3 GB Tight60
16-bit (F16/BF16)29.0 GB No29.7 GB No34.2 GB No—

• real GGUF file size. Speed is a bandwidth ceiling (936 GB/s), not a benchmark. Methodology.

More

Qwen2.5 Coder 14B Instruct on every GPU and Mac · Every model that runs on the RTX 3090 · Other models that fit 24 GB VRAM · Try other settings in the calculator

Data: Hugging Face Hub (Qwen/Qwen2.5-Coder-14B-Instruct), NVIDIA specs. Data as of 2026-11-17.