How much memory does gpt-oss-20b need at Q4?

The Q4 figure used here is 11.8 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 gpt-oss-20b at Q4?

Radeon RX 7600 XT, list price $329.

How fast is gpt-oss-20b on the Radeon RX 7600 XT, as a bandwidth estimate?

About 138 tok/s at Q8. That is device bandwidth divided by active weight bytes, not a measured run.

Yes. gpt-oss-20b fits on the Radeon RX 7600 XT at Q8, with 2.9 GB of headroom. Usable memory is 15.2 GB.

Can the Radeon RX 7600 XT run gpt-oss-20b?

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 gpt-oss-20b on Radeon RX 7600 XT
Quant	Estimate	Headroom	Est. tok/s	Verdict
Q4	11.8 GB	3.4 GB	144 tok/s	Yes
Q8	12.3 GB	2.9 GB	138 tok/s	Yes
FP16	50.2 GB	-35.0 GB	40.0 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: 0.2 GB at 8,192 tokens and 0.8 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 128 tokens. The figures above assume every counted layer stores the full context, so they can overstate cache on sliding layers.

Offload

The Q4 weight estimate fits in usable VRAM, so this page does not assume a system-RAM offload.

Ollama

This command pulls the Q4_K_M file from unsloth/gpt-oss-20b-GGUF, the Hub repo whose file size is used above. It is not an official Ollama library tag.

ollama run hf.co/unsloth/gpt-oss-20b-GGUF:Q4_K_M

Other picks

Cheapest catalog device at Q4: Radeon RX 7600 XT
gpt-oss-20b is the catalog model whose Q4 estimate sits closest under this device's usable memory.

All hardware for gpt-oss-20b · All models for the Radeon RX 7600 XT

Questions

How much memory does gpt-oss-20b need at Q4?: The Q4 figure used here is 11.8 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 7600 XT run gpt-oss-20b?: Yes. gpt-oss-20b fits on the Radeon RX 7600 XT at Q8, with 2.9 GB of headroom. Usable memory is 15.2 GB.
What is the cheapest catalog device that runs gpt-oss-20b at Q4?: Radeon RX 7600 XT, list price $329.
How fast is gpt-oss-20b on the Radeon RX 7600 XT, as a bandwidth estimate?: About 138 tok/s at Q8. That is device bandwidth divided by active weight bytes, not a measured run.