RunsOnMy ModelsGPUs & MacsBy VRAMBest GPUCan I run…?OllamaStart hereAPI pricesVRAM calculatorNew

Can Gemma 4 31B IT run on the Arc A770 16GB?

Yes. At 8K context the best fit is Q2_K, using about 14.7 GB of the 15.2 GB usable on the Arc A770 16GB. Estimated speed ceiling: ~44 tokens/sec.

Better quality with offload: 4-bit (Q4_K_M)

Runs with offload 4-bit (Q4_K_M) at 8K context: 40 of 60 layers on the Arc A770 16GB (about 14.4 GB) and the other 20 in system RAM (about 6.8 GB). Needs a PC with at least 32 GB of system RAM. Estimated speed: 3.2–5.9 tokens/sec (slow but usable), versus a bandwidth ceiling of ~29 tok/s if it fit fully in VRAM.

llama-server -hf bartowski/google_gemma-4-31B-it-GGUF:Q4_K_M -c 8192 -ngl 40 --no-mmproj

Estimate, not a benchmark. The command downloads the GGUF from Hugging Face. The speed range spans dual-channel DDR4-3200 to DDR5-6000 system RAM; your CPU, RAM speed and llama.cpp version move it. If llama.cpp runs out of GPU memory, lower -ngl by 2-3; if VRAM is left over, raise it for more speed. Recent llama.cpp builds can also pick the split automatically. How we estimate offload.

+ 1.1 GB with image input. The figures on this page are for text only. With image input enabled, the vision projector (mmproj) file loads alongside the model: Q2_K at 8K context then needs about 15.8 GB (no longer fits, so drop one quant or shorten the context). llama.cpp's -hf downloads and loads it automatically unless you add --no-mmproj. Calculator with image input on.

Memory by quant and context

Quant4K ctx8K ctx32K ctx128K ctxMax tok/s (est., 8K)
2-bit (Q2_K) •14.1 GB Tight14.7 GB Tight18.4 GB No33.4 GB No44
3-bit (Q3_K_M) •17.1 GB No17.8 GB No21.5 GB No36.5 GB No—
4-bit (Q4_K_M) •20.6 GB No21.2 GB No24.9 GB No39.9 GB No—
5-bit (Q5_K_M) •23.4 GB No24.0 GB No27.7 GB No42.7 GB No—
6-bit (Q6_K) •27.2 GB No27.8 GB No31.6 GB No46.6 GB No—
8-bit (Q8_0) •32.7 GB No33.3 GB No37.1 GB No52.1 GB No—
16-bit (F16/BF16) •59.5 GB No60.1 GB No63.9 GB No78.9 GB No—

• real GGUF file size from bartowski/google_gemma-4-31B-it-GGUF. Speed is a bandwidth ceiling (560 GB/s), not a benchmark. Methodology.

What fits instead

"Runs" and "fits" mean the whole model in GPU memory at 8K context (4-bit or better for the model pick). Prices are US launch MSRPs, not today's street prices. "Popular" means the 60 most-downloaded models we track.

More

Gemma 4 31B IT on every GPU and Mac · Every model that runs on the Arc A770 16GB · Other models that fit 16 GB VRAM · Try other settings in the calculator

Guides: Best GPU for local LLMs by budget · Can I run gpt-oss, Llama 70B or Qwen3? · Mac vs PC for local LLMs

Data: Hugging Face Hub (google/gemma-4-31B-it), Intel specs. Model data last checked 2026-11-18; refreshed hourly.