Yes — with 2.8 GB to spare

gpt-oss 20B at MXFP4 fits your RTX A4000 entirely on the GPU at 8K context, at an estimated 56 tokens per second. Past 126K the KV cache pushes it over — quantise the cache to q8_0, or step down a quantisation, to go longer.

Fully on GPU 8K context MXFP4 · 10.8 GB Apache 2.0 Released Aug 2025 Native MXFP4

Ships natively in MXFP4, so the 4-bit weights are the reference weights, not a lossy copy. Fits 16 GB.

What hardware do I need for gpt-oss 20B? →

The VRAM budget

weights 10.8 GB

Weights 10.8 GB KV cache @ 8K 0.19 GB Runtime overhead 0.6 GB Free 2.8 GB of 14.4 GB

Quantisation ladder

Quant	Weights	Total @ 8K	Max context	Tok/s	Quality	Fit
MXFP4	10.8 GB	11.6 GB	126K	56	Reference	Recommended

Quality is the published perplexity delta against f16 weights. Max context assumes an f16 KV cache; q8_0 roughly doubles it. This model interleaves sliding-window layers (128 tokens, 1 global in 2), which is why its cache barely grows with context.

How to run it

Ollama llama.cpp LM Studio vLLM

terminal

$ ollama pull gpt-oss:20b
$ OLLAMA_CONTEXT_LENGTH=8192 \
    ollama run gpt-oss:20b

The default. One binary, a model registry, an OpenAI-compatible port. More on Ollama.

01Download is 10.8 GB. Keep it on an SSD — a first load off a spinning disk takes minutes.

02Close anything else holding VRAM. A browser with hardware acceleration can sit on 1–2 GB.

03There is room to go to 126K context on this card.

See all models for this rig Compare with another model