Model benchmarks

openai/gpt-oss-120b local LLM performance

As of August 2026, openai/gpt-oss-120b runs at up to 176.4 tok/s for local inference (best of 2 community benchmark runs across 2 GPUs).

LM StudioOllamaMXFP4
ShareRedditX

Model size

120B

Peak speed

176.4 tok/s

Average speed

111.8 tok/s

Min memory

59.0 GB

Max context

65,536 tokens

Avg output / run

9,119 tokens

Avg runtime / run

3m 31s

Avg quality

79.9

Benchmark runs

2

GPUs tested

2

Quality by task

Average LLM-judged quality (0–100) with the run-to-run spread shown as a P5–P95 band, overall and for each benchmark task, across all 2 runs. The low and high columns show how much the judge’s score varies between runs, and need at least two runs to display.

TaskP5 (low)AvgP95 (high)
Overall79.779.980.2
Agent Workflow83.786.388.8
Code Generation74.074.174.2
Role Play & Narrative78.680.782.7
Research & Analysis78.478.879.1

Performance by hardware and tool

Every hardware/tool/quantization combination openai/gpt-oss-120b has been benchmarked on, ranked by peak token generation speed. Last updated August 2026.

HardwareToolQuantBest tok/sAvg tok/sMemoryContextQualityRuns
NVIDIA L4OllamaMXFP4176.4 tok/s176.4 tok/s60.0 GB65,536 tokens79.71
Apple M2 MaxLM Studio47.1 tok/s47.1 tok/s59.0 GB8,192 tokens80.21

Frequently asked questions

Is gpt-oss:120b good for coding?
In our benchmarks, gpt-oss:120b scores 74.1/100 for coding. It runs at about 111.8 tok/s, so if you want more speed, Gemma4:E2B/QAT-MTP@131K is faster (~303.9 tok/s) and still scores well for coding (66.9/100).
Is gpt-oss:120b good for agentic (tool-using) tasks?
In our benchmarks, gpt-oss:120b scores 86.3/100 for agentic workflows. It runs at about 111.8 tok/s, so if you want more speed, mudler/Ornith-1.5-35B-A3B-APEX-MTP-GGUF is faster (~190.9 tok/s) and still scores well for agentic workflows (85.7/100).
How fast is openai/gpt-oss-120b for local inference?
Across 2 community benchmark runs, openai/gpt-oss-120b reaches up to 176.4 tok/s and averages 111.8 tok/s, with the fastest results on NVIDIA L4.
How much memory does openai/gpt-oss-120b need?
The leanest observed configuration used about 59.0 GB of memory (quantizations tested: MXFP4).
Which tools have been used to run openai/gpt-oss-120b?
Benchmarks were submitted using LM Studio, Ollama. Results are community-contributed and updated as new runs arrive.