gpt-oss-120b

high

Creator

OpenAI

Released

2025-08-05

Intelligence

23.8

Artificial Analysis Index

Coding

30.4

Artificial Analysis Index

In $/1M

$0.15

input tokens

Out $/1M

$0.60

output tokens

Blended $/1M

$0.26

3:1 blended

Speed

193

tokens / sec

Profile

License, openness, and modality — the governance layer.

License

Apache 2.0

commercial OK

Weights

Open

downloadable

Modalities

Text

Parameters

117B

total

GPT-OSS-120B is an open-weight, 116.8B-parameter Mixture-of-Experts (MoE) language model from OpenAI designed for high-reasoning, agentic, and general-purpose production use cases. It activates 5.1B parameters per forward pass and is optimized to run on a single H100 GPU with native MXFP4 quantization. The model supports configurable reasoning depth, full chain-of-thought access, and native tool use, including function calling, browsing, and structured output generation. It achieves near-parity with OpenAI o4-mini on core reasoning benchmarks. Note: While referred to as '120b' for simplicity, it technically has 116.8B parameters.

Capability profile

Category strength across 20 domains, via LLM Stats.

legal
90finance
90language
90biology
80physics
80chemistry
80reasoning
70
tool calling
70communication
70math
60general
60healthcare
60vision
10
Via LLM Stats

Benchmark breakdown

Independent evaluation scores, normalized to 0–100.

GPQA Diamond
78
Humanity’s Last Exam
19
MMLU-Pro
81
SciCode
39
LiveCodeBench
88
MATH-500
AIME 2025
93
τ²-Bench (agentic)
66
Terminal-Bench Hard
23
IFBench
69
Via Artificial Analysis
© 2026 NYSGPT2525 LLC