RunsOnMy ModelsGPUs & MacsBy VRAMAPI pricesVRAM calculatorNew

Can Qwen2.5 Coder 14B Instruct run on the RTX 3070?

Yes. At 8K context the best fit is Q2_K, using about 7.3 GB of the 7.6 GB usable on the RTX 3070. Estimated speed ceiling: ~81 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 Tight11.8 GB No81
3-bit (Q3_K_M)8.2 GB No8.9 GB No13.4 GB No—
4-bit (Q4_K_M)9.8 GB No10.6 GB No15.1 GB No—
5-bit (Q5_K_M)11.2 GB No12.0 GB No16.5 GB No—
6-bit (Q6_K)12.7 GB No13.5 GB No18.0 GB No—
8-bit (Q8_0)16.1 GB No16.8 GB No21.3 GB No—
16-bit (F16/BF16)29.0 GB No29.7 GB No34.2 GB No—

• real GGUF file size. Speed is a bandwidth ceiling (448 GB/s), not a benchmark. Methodology.

More

Qwen2.5 Coder 14B Instruct on every GPU and Mac · Every model that runs on the RTX 3070 · Other models that fit 8 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.