Can I Run Gemma 4 on RTX 4090 Desktop (24 GB VRAM, 64 GB RAM)?

Written by Jakub Rusinowski · Last updated June 3, 2026

Yes

Yes — Gemma 4 31B at Q4_K_M needs about 21.3 GB of the 24 GB usable on RTX 4090 Desktop (24 GB VRAM, 64 GB RAM) (~2.7 GB spare), at ~35.5 tok/s (estimated), with room for about 16,384 tokens of context.

Confidence: medium · Recommended quantization: Q4_K_M · Estimated speed: ~35.5 tok/s

RTX 4090 Desktop (24 GB VRAM, 64 GB RAM) — what it gives a model

Usable memory for models24 GB
Memory bandwidth1008 GB/s
Form factorDesktop
Operating systemWindows or Linux
Memory upgradeableYes

Gemma 4 on RTX 4090 Desktop (24 GB VRAM, 64 GB RAM): memory by quantization

QuantMemory neededFits 24 GB?Max contextEst. speedDownload
F1664.6 GB✗ No62 GB
Q8_035.5 GB✗ No32.9 GB
Q6_K28 GB✗ No25.4 GB
Q5_K_M24.6 GB✗ No22 GB
Q4_K_M21.3 GB✓ Yes16K~35.5 tok/s18.7 GB
Q3_K_M15.8 GB✓ Yes32K~47.8 tok/s13.2 GB
Q2_K12.8 GB✓ Yes32K~59.1 tok/s10.2 GB

Which Gemma 4 sizes fit

VariantNeeds at Q4_K_MFits?Est. speed
Gemma 4 31B21.3 GB✓ Fits~35.5 tok/s
Gemma 4 26B-A4B18.2 GB✓ Fits~152.2 tok/s
Gemma 4 12B (Unified)9.4 GB✓ Fits~78.8 tok/s
Gemma 4 E4B6.8 GB✓ Fits~106.5 tok/s
Gemma 4 E2B4.9 GB✓ Fits~143.2 tok/s

What to watch out for

RTX 4090 desktop limitations

Recommended setup

Ollama or llama.cpp (CUDA/ROCm) — vLLM if you need concurrent serving

How these numbers are calculated

FAQ

Can I run Gemma 4 on RTX 4090 Desktop (24 GB VRAM, 64 GB RAM)?

Yes — Gemma 4 31B at Q4_K_M needs about 21.3 GB of the 24 GB usable on RTX 4090 Desktop (24 GB VRAM, 64 GB RAM) (~2.7 GB spare), at ~35.5 tok/s (estimated), with room for about 16,384 tokens of context.

Which quantization of Gemma 4 should I use on RTX 4090 Desktop (24 GB VRAM, 64 GB RAM)?

Q4_K_M — it needs about 21.3 GB of the 24 GB available, downloads as roughly 18.7 GB, and runs at an estimated 35.5 tokens/sec with up to 16K of context.

What limits Gemma 4 on RTX 4090 Desktop (24 GB VRAM, 64 GB RAM)?

Nothing binding — the model fits with headroom and generates at a usable speed on this hardware.

Which runtime should I use?

Ollama or llama.cpp (CUDA/ROCm) — vLLM if you need concurrent serving

Other Computers

Other Models on RTX 4090 Desktop (24 GB VRAM, 64 GB RAM)

Gemma 4 on GPUs

What This Model Is Good At

Model & Tools

← Can I Run It? | Gemma 4 model page | Check your hardware