Venice Atlas
← Back to model catalog

VENICE ATLAS · MODEL DOSSIER · OPENAI GPT OSS 120B

Model overview

OpenAI GPT OSS 120B

gpt-oss-120b is an open-weight, 117B-parameter Mixture-of-Experts (MoE) language model from OpenAI designed for high-reasoning, agentic, and general-purpose production use cases. The model supports configurable reasoning depth, full chain-of-thought access, and native tool use, including function calling, browsing, and structured output generation

openai-gpt-oss-120b · venice.ai
Text · Venice catalog$0.07 per 1M tokensPrice on Venice AI
ModalityTextOn Venice AI
Primary price$0.07 per 1M tokensper 1M tokens
Context window128KMax tokens
Benchmarks7Benchmark results

Benchmark results

See the tests behind each score. They use the benchmark’s original scoring scale.

Healthcare#1830.26 rating
Chat1 result
TAU-bench Retailtau-bench-retail · chat
67.8%
ReportedOct 2026
General intelligence1 result
MMLUmmlu · general
90.0%
ReportedOct 2026
Healthcare2 results
HealthBenchhealthbench · healthcare
57.6%
ReportedOct 2026
HealthBench Hardhealthbench-hard · healthcare
30.0%
ReportedOct 2026
Math1 result
CodeForcescodeforces · math
82.1%
ReportedOct 2026
Reasoning2 results
GPQAgpqa · reasoning
80.1%
ReportedOct 2026
Humanity's Last Examhumanity's-last-exam · reasoning
14.9%
ReportedOct 2026

Learn what each benchmark measures ↗

Related models

More from openai

openai-gpt-4o-2024-11-20GPT-4o$2.50 input per 1M tokens ↗openai-gpt-52GPT-5.2$1.75 input per 1M tokens ↗openai-gpt-53-codexGPT-5.3 Codex$1.75 input per 1M tokens ↗