Can I run it?
Can I run gemma 4 31B it on GeForce RTX 5070?
Heavily degraded
Heavily degraded fully on your GeForce RTX 5070 at UD-IQ2_XXS, 8K context, about 22–30 tok/s. Weights 7.95 GB plus 2.03 GB of KV cache — 10.4 GB of 10.9 GB usable.
Best quantUD-IQ2_XXS
Working context8K
Max that fits8K
Speed22–30 tok/sEstimate from memory bandwidth, not a benchmark. Method.
Weights7.95 GB
KV cache2.03 GB
Total10.4 GB
Usable10.9 GB
gemma 4 31B it details · GeForce RTX 5070 page · How this estimate is made
Run commands
llama.cpp
llama-server -hf unsloth/gemma-4-31B-it-GGUF:UD-IQ2_XXS -c 8192 -ngl 60
Ollama
ollama run gemma
LM Studio
lms get unsloth/gemma-4-31B-it-GGUF # Then load quant UD-IQ2_XXS
Assumptions
- Estimate from memory bandwidth, not a benchmark. Decode speed uses 55–75% of peak bandwidth.
- 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-IQ2_XXS 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.
- This quant is IQ1 or IQ2. It can load, but quality is heavily degraded, so it is left off best-fit lists.
Larger models that also fit
- Can I run Qwen3.6 35B A3B on GeForce RTX 5070? — Heavily degraded, UD-IQ2_XXS, 81–111 tok/s (estimate)
- Can I run Qwen3.5 35B A3B on GeForce RTX 5070? — Heavily degraded, UD-IQ2_XXS, 82–112 tok/s (estimate)
- Can I run Qwen2.5 32B Instruct on GeForce RTX 5070? — Heavily degraded, IQ2_XXS, 37–50 tok/s (estimate)
Updated 2026-10-24. Profile on this page is the device default plus 32 GB of DDR5-5600 system RAM.