Model benchmarks

openai/gpt-oss-20b local LLM performance

As of April 2026, openai/gpt-oss-20b runs at up to 111.1 tok/s for local inference (best of 9 community benchmark runs across 1 GPU).

LM Studio

Model size

20B

Peak speed

111.1 tok/s

Average speed

109.1 tok/s

Min memory

9.8 GB

Max context

256,000 tokens

Best quality

89.2

Benchmark runs

9

GPUs tested

1

Performance by hardware and tool

Every hardware/tool/quantization combination openai/gpt-oss-20b has been benchmarked on, ranked by peak token generation speed. Last updated April 2026.

HardwareToolQuantBest tok/sAvg tok/sMemoryContextQualityRuns
AMD Radeon RX 7900 XTXLM Studio111.1 tok/s109.1 tok/s9.8 GB256,000 tokens66.89

Frequently asked questions

How fast is openai/gpt-oss-20b for local inference?
Across 9 community benchmark runs, openai/gpt-oss-20b reaches up to 111.1 tok/s and averages 109.1 tok/s, with the fastest results on AMD Radeon RX 7900 XTX.
How much memory does openai/gpt-oss-20b need?
The leanest observed configuration used about 9.8 GB of memory.
Which tools have been used to run openai/gpt-oss-20b?
Benchmarks were submitted using LM Studio. Results are community-contributed and updated as new runs arrive.