How much memory does Gemma 4 26B-A4B need at Q4?

The Q4 figure used here is 17.3 GB. GGUF file size plus an FP16 KV-cache estimate at 8,192 tokens. The cache counts only the layers and heads named in the method on the About page.

What is the cheapest catalog device that runs Gemma 4 26B-A4B at Q4?

Radeon RX 7900 XT, list price $899.

How fast is Gemma 4 26B-A4B on the Radeon RX 9070 XT, as a bandwidth estimate?

About 256 tok/s at Q4. That is device bandwidth divided by active weight bytes, not a measured run.

No. Gemma 4 26B-A4B does not fit on the Radeon RX 9070 XT at Q4. The estimate is 17.3 GB and usable memory is 15.2 GB.

Can the Radeon RX 9070 XT run Gemma 4 26B-A4B?

Data updated 16 October 2026. Last reviewed the same day, when this catalog snapshot was written.

Usable memory on this 16 GB VRAM config is 15.2 GB (95% of the sticker figure). A result under 2 GB of headroom is labeled tight.

Quant ladder

Estimated memory for Gemma 4 26B-A4B on Radeon RX 9070 XT
Quant	Estimate	Headroom	Est. tok/s	Verdict
Q4	17.3 GB	-2.1 GB	256 tok/s	No
Q8	27.2 GB	-12.0 GB	162 tok/s	No
FP16	61.9 GB	-46.7 GB	84.2 tok/s	No

GGUF file size plus an FP16 KV-cache estimate at 8,192 tokens. The cache counts only the layers and heads named in the method on the About page.
Formula estimate: parameter count × bytes per weight × 1.20. No GGUF file size is stored for this quant.
Tokens/sec is a bandwidth roofline: GB/s divided by active weight GB. It is not a measured generation speed.

KV cache at 8K and 32K

FP16 KV-cache estimate: 0.3 GB at 8,192 tokens and 1.3 GB at 32,768 tokens. GQA counts two tensors (K and V) for the stored KV heads and head dimension. config.json also sets a sliding window of 1,024 tokens. The figures above assume every counted layer stores the full context, so they can overstate cache on sliding layers.

Offload

About 1.7 GB of the Q4 weights sit outside usable VRAM. A partial offload would need at least that much free system RAM. This catalog does not record system RAM, so offload is not counted as a fit.

Ollama

This command pulls the Q4_K_M file from unsloth/gemma-4-26B-A4B-it-GGUF, the Hub repo whose file size is used above. It is not an official Ollama library tag.

ollama run hf.co/unsloth/gemma-4-26B-A4B-it-GGUF:Q4_K_M

What fits instead

Cheapest catalog device at Q4: Radeon RX 7900 XT
Closest catalog model that still fits at Q4: gpt-oss-20b.

All hardware for Gemma 4 26B-A4B · All models for the Radeon RX 9070 XT

Questions

How much memory does Gemma 4 26B-A4B need at Q4?: The Q4 figure used here is 17.3 GB. GGUF file size plus an FP16 KV-cache estimate at 8,192 tokens. The cache counts only the layers and heads named in the method on the About page.
Can the Radeon RX 9070 XT run Gemma 4 26B-A4B?: No. Gemma 4 26B-A4B does not fit on the Radeon RX 9070 XT at Q4. The estimate is 17.3 GB and usable memory is 15.2 GB.
What is the cheapest catalog device that runs Gemma 4 26B-A4B at Q4?: Radeon RX 7900 XT, list price $899.
How fast is Gemma 4 26B-A4B on the Radeon RX 9070 XT, as a bandwidth estimate?: About 256 tok/s at Q4. That is device bandwidth divided by active weight bytes, not a measured run.