Can I run Llama 3.2 1B on AMD RX 7900 XTX 24GB?
Yes: Llama 3.2 1B runs on AMD RX 7900 XTX 24GB with F16 at roughly 227 tokens/s. Estimated memory 3.2 GiB at 8,192 tokens of context.
With F16 you can push the context to about 131,072 tokens before memory runs out (F16 KV cache).
Memory by quantisation (8,192-token context)
| Quantisation | Memory | Verdict | Est. speed |
|---|---|---|---|
| F16Unquantized reference | 3.2 GiB | Fits | ~227 tok/s |
| Q8_0Near-lossless | 2.2 GiB | Fits | ~393 tok/s |
| Q6_KVery close to F16 | 1.9 GiB | Fits | ~486 tok/s |
| Q5_K_MGreat quality/size | 1.8 GiB | Fits | ~542 tok/s |
| Q4_K_MBest default for most | 1.6 GiB | Fits | ~612 tok/s |
| IQ4_XSSmaller 4-bit (imatrix) | 1.6 GiB | Fits | ~673 tok/s |
| Q3_K_MNoticeable loss | 1.5 GiB | Fits | ~714 tok/s |
| Q2_KLast resort; big loss | 1.4 GiB | Fits | ~792 tok/s |
AMD RX 7900 XTX 24GB offers about 24 GiB of memory (92% usable for a model) at 960 GB/s. Speed is a bandwidth-bound estimate of single-stream generation at 65% efficiency; real results vary by runtime, drivers and settings, so measure on your machine.
How this is calculated
Weights ≈ parameters × bits per weight ÷ 8. KV cache ≈ 2 × 16 layers × 8 KV heads × 64 head dim × 2 bytes per token. Decode speed ≈ efficiency × memory bandwidth ÷ bytes read per token. Learn the maths in VRAM math and memory bandwidth, or try the VRAM calculator.
Llama 3.2 1B on other hardware
Similar models on AMD RX 7900 XTX 24GB
Running slowly or out of memory? See CUDA out of memory and the quantisation lesson.