Can I run Llama 3.2 1B on NVIDIA RTX 4090 24GB?
Yes: Llama 3.2 1B runs on NVIDIA RTX 4090 24GB with F16 at roughly 238 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 | ~238 tok/s |
| Q8_0Near-lossless | 2.2 GiB | Fits | ~413 tok/s |
| Q6_KVery close to F16 | 1.9 GiB | Fits | ~510 tok/s |
| Q5_K_MGreat quality/size | 1.8 GiB | Fits | ~570 tok/s |
| Q4_K_MBest default for most | 1.6 GiB | Fits | ~642 tok/s |
| IQ4_XSSmaller 4-bit (imatrix) | 1.6 GiB | Fits | ~707 tok/s |
| Q3_K_MNoticeable loss | 1.5 GiB | Fits | ~749 tok/s |
| Q2_KLast resort; big loss | 1.4 GiB | Fits | ~832 tok/s |
NVIDIA RTX 4090 24GB offers about 24 GiB of memory (94% usable for a model) at 1008 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 4090 24GB
Running slowly or out of memory? See CUDA out of memory and the quantisation lesson.