RunsOnMy ModelsGPUs & MacsBy VRAMAPI pricesVRAM calculatorNew

Can falcon 40b run on the RX 7900 XT?

Yes. At 8K context the best fit is Q2_K, using about 16.1 GB of the 19.0 GB usable on the RX 7900 XT. Estimated speed ceiling: ~51 tokens/sec.

Memory by quant and context

Quant4K ctx8K ctx32K ctx128K ctxMax tok/s (est., 8K)
2-bit (Q2_K) •15.7 GB Fits16.1 GB Fits19.0 GB Tight30.2 GB No51
3-bit (Q3_K_M) •19.7 GB No20.2 GB No23.0 GB No34.2 GB No—
4-bit (Q4_K_M) •24.7 GB No25.2 GB No28.0 GB No39.3 GB No—
5-bit (Q5_K_M) •29.5 GB No30.0 GB No32.8 GB No44.1 GB No—
6-bit (Q6_K) •33.1 GB No33.6 GB No36.4 GB No47.6 GB No—
8-bit (Q8_0)42.4 GB No42.9 GB No45.7 GB No56.9 GB No—
16-bit (F16/BF16)78.9 GB No79.4 GB No82.2 GB No93.4 GB No—

• real GGUF file size from mradermacher/falcon-40b-i1-GGUF. Speed is a bandwidth ceiling (800 GB/s), not a benchmark. Methodology.

More

falcon 40b on every GPU and Mac · Every model that runs on the RX 7900 XT · Other models that fit 24 GB VRAM · Try other settings in the calculator

Data: Hugging Face Hub (tiiuae/falcon-40b), AMD specs. Data as of 2026-11-17.