Rent a GPU

Can the AMD Radeon RX 7900 XT run Gemma 4 31B?

Radeon RX 7900 XT × Gemma 4 31B8K context
Yes, with heavy quantizationFits only at a low-quality quant
Memory at Q3_K_M17.1 GB of 20 GB
Weights 15.3 GBKV cache 1.2 GBRuntime 0.6 GB
QuantQ3_K_MNoticeably degraded
Speed~31tokens per second
Max context77Ktokens that fit at Q3_K_M
  1. FP16
  2. Q8_0
  3. Q6_K
  4. Q5_K_M
  5. Q4_K_M
  6. Q3_K_M
  7. Q2_K

Only just. Gemma 4 31B fits on the AMD Radeon RX 7900 XT at Q3_K_M, which costs noticeable quality. Expect about 31 tokens/second. A card with more memory is a better experience.

Try other context lengths

Run it with Ollama

Ollama's default gemma4:31b download is Q4_K_M, which needs about 20.8 GB, more than the 20 GB available. Download the Q3_K_M GGUF instead (LM Studio lists quants per model), or try a smaller context length.

New to this? Step-by-step setup for the Radeon RX 7900 XT

Run it in the cloud

Want full quality? Run Gemma 4 31B on a bigger rented GPU.

Rent a cloud GPU by the hour. No upfront cost, stop any time. Not sure whether to rent or buy? Compare the cost.

Rent a GPU on RunPodRent a GPU on Vast.ai

Memory needed by quantization

At 8,192 tokens of context. The black line marks the 20 GB the AMD Radeon RX 7900 XT can use.

QuantQualityMemoryFits?Speed
FP16Full quality64.4 GBToo big—
Q8_0Near-lossless35 GBToo big—
Q6_KExcellent27.4 GBToo big—
Q5_K_MVery good24 GBToo big—
Q4_K_MGood, most popular20.8 GBToo big—
Q3_K_MNoticeably degraded17.1 GBFits~31 tok/s
Q2_KHeavily degraded13.5 GBFits~41 tok/s

Measured speed from real Radeon RX 7900 XT owners

Tokens per second that people measured on their own AMD Radeon RX 7900 XT, next to our estimate.

No one has measured Gemma 4 31B on the AMD Radeon RX 7900 XT yet. The speeds above are estimates. Run the benchmark and yours becomes the first real number on this page.

Have a Radeon RX 7900 XT? Add your real speed

One command times Gemma 4 31B in your own Ollama, shows you exactly what it would send, and asks before sending anything. No sign-up, no personal data.

macOS and Linux

curl -fsSL https://gpufit.dev/bench.sh | sh -s -- gemma4:31b

Windows (PowerShell)

$env:GPUFIT_MODEL="gemma4:31b"; irm https://gpufit.dev/bench.ps1 | iex

How the benchmark works

GPUs that run Gemma 4 31B comfortably

Check RTX 3090 prices

Models that run well on the AMD Radeon RX 7900 XT

Frequently asked questions

Can AMD Radeon RX 7900 XT run Gemma 4 31B?

Only just. Gemma 4 31B fits on the AMD Radeon RX 7900 XT at Q3_K_M, which costs noticeable quality. Expect about 31 tokens/second. A card with more memory is a better experience.

How much VRAM does Gemma 4 31B need?

About 20.8 GB at Q4_K_M with 8,192 tokens of context, and about 64.4 GB at full FP16 precision.

How fast is Gemma 4 31B on the AMD Radeon RX 7900 XT?

Roughly 31 tokens per second at Q3_K_M, estimated from the AMD Radeon RX 7900 XT's 800 GB/s memory bandwidth.