Can I run Mixtral 8x7B (MoE) on NVIDIA RTX 4090 24GB?
Yes: Mixtral 8x7B (MoE) runs on NVIDIA RTX 4090 24GB with Q2_K at roughly 101 tokens/s. Estimated memory 20.8 GiB at 8,192 tokens of context.
Memory by quantisation (8,192-token context)
| Quantisation | Memory | Verdict | Est. speed |
|---|---|---|---|
| F16Unquantized reference | 89.6 GiB | Does not fit | — |
| Q8_0Near-lossless | 48.8 GiB | Needs CPU offload | — |
| Q6_KVery close to F16 | 38.3 GiB | Needs CPU offload | — |
| Q5_K_MGreat quality/size | 33.5 GiB | Needs CPU offload | — |
| Q4_K_MBest default for most | 29.0 GiB | Needs CPU offload | — |
| IQ4_XSSmaller 4-bit (imatrix) | 25.7 GiB | Needs CPU offload | — |
| Q3_K_MNoticeable loss | 23.9 GiB | Needs CPU offload | — |
| Q2_KLast resort; big loss | 20.8 GiB | Tight | ~101 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 × 32 layers × 8 KV heads × 128 head dim × 2 bytes per token. Decode speed ≈ efficiency × memory bandwidth ÷ bytes read per token (this is a mixture-of-experts model, so only the active parameters are read, but all of them must fit in memory). Learn the maths in VRAM math and memory bandwidth, or try the VRAM calculator.
Mixtral 8x7B (MoE) 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.