Can I run it?
Can I run gemma 4 31B it on GeForce RTX 5090?
Tight fit
Tight fit fully on your GeForce RTX 5090 at UD-Q6_K_XL, 8K context, about 28–38 tok/s. Weights 25.6 GB plus 2.03 GB of KV cache — 28.2 GB of 29.9 GB usable.
Best quantUD-Q6_K_XL
Working context8K
Max that fits8K
Speed28–38 tok/s
Weights25.6 GB
KV cache2.03 GB
Total28.2 GB
Usable29.9 GB
gemma 4 31B it details · GeForce RTX 5090 page · How this estimate is made
Run commands
llama.cpp
llama-server -hf unsloth/gemma-4-31B-it-GGUF:UD-Q6_K_XL -c 8192 -ngl 60
Ollama
ollama run gemma
LM Studio
lms get unsloth/gemma-4-31B-it-GGUF # Then load quant UD-Q6_K_XL
Assumptions
- Estimate, not a benchmark. Decode speed uses a memory-bandwidth model at 55–75% of peak.
- KV cache is FP16. Engine overhead assumes llama.cpp (0.3 GB + 1% of the weights). Ollama is closer to 0.9 GB + 3%, LM Studio closer to 1 GB + 3%.
- UD-Q6_K_XL size is the real GGUF file size from Hugging Face.
- 0.5 GB is reserved because this GPU also drives a display, then 5% of the rest is left free.
- Speed is quoted at 8K context, not at the maximum context that still fits.
Larger models that also fit
- Can I run Qwen3.6 35B A3B on GeForce RTX 5090? — Tight fit, UD-Q6_K, 364–497 tok/s
- Can I run Qwen3.8 35B A3B Distill on GeForce RTX 5090? — Tight fit, Q6_K, 374–510 tok/s
- Can I run Qwen3.5 35B A3B on GeForce RTX 5090? — Tight fit, UD-Q6_K, 374–510 tok/s
Updated 2026-10-24. Profile on this page is the device default plus 32 GB of DDR5-5600 system RAM.