Can I run Llama 3.2 1B on NVIDIA RTX 5090 32GB?
Yes: Llama 3.2 1B runs on NVIDIA RTX 5090 32GB with F16 at roughly 424 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 | ~424 tok/s |
| Q8_0Near-lossless | 2.2 GiB | Fits | ~734 tok/s |
| Q6_KVery close to F16 | 1.9 GiB | Fits | ~906 tok/s |
| Q5_K_MGreat quality/size | 1.8 GiB | Fits | ~1,013 tok/s |
| Q4_K_MBest default for most | 1.6 GiB | Fits | ~1,142 tok/s |
| IQ4_XSSmaller 4-bit (imatrix) | 1.6 GiB | Fits | ~1,256 tok/s |
| Q3_K_MNoticeable loss | 1.5 GiB | Fits | ~1,332 tok/s |
| Q2_KLast resort; big loss | 1.4 GiB | Fits | ~1,479 tok/s |
NVIDIA RTX 5090 32GB offers about 32 GiB of memory (94% usable for a model) at 1792 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 NVIDIA RTX 5090 32GB
Running slowly or out of memory? See CUDA out of memory and the quantisation lesson.