NOUMENOpen app
NOUMEN

OpenAI · Open weights

GPT OSS 120B

Added Feb 3, 2026Open weights

Compare models

Choose up to 3 models to compare.0 of 3 chosen

Loading models…

Use in the app

An open-weight, 117B-parameter Mixture-of-Experts (MoE) language model 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.

  • Reasoning
  • Tools
  • Structured output
Input
Text
Output
Text
Discovery tier
Open weights

Pricing

USD per 1M tokens
Input
$0.03
Output
$0.15
Cache read
$0.03

Reference prices as published by the model provider. API example

Specifications

Context window
128K
Max output
16.4K
Average output (7 days)
1.1K tokens
Longer than 74% of 468 models

Side by side

Compare models

Put GPT OSS 120B next to up to three other models. The comparison stays in the link, so you can share it.

Compare models

Choose up to 3 models to compare.0 of 3 chosen

Loading models…

TryMiniMax M1 80KMistral Small 4 119B ThinkingQwen 3 Coder 480BAll 3

Public evaluations

Benchmarks

Independent evaluation results, with how each score ranks among the models in the Noumen catalog.

Intelligence Index

11.6

Tested at high reasoning effort

Better than 25% of 326 scored models

3.8Catalog models by score55.1

Indexes

  • Coding Index

    30.4

    Composite of the coding evaluations

    Better than 22% of 168 scored models

  • Math Index

    93.4

    Composite of the math evaluations

    Better than 93% of 118 scored models

Agentic work

  • AutomationBench-AA

    0.2%

    Workflow automation, penalized for guardrail violations

    Better than 2% of 173 scored models

  • AutomationBench-AA tasks completed

    0.0%

    Workflows finished end to end without a guardrail violation

    Better than 0% of 173 scored models

  • GDPval-AAv2.1

    608 Elo

    Economically valuable occupational tasks, rated head to head

    Better than 16% of 187 scored models

Document reasoning

  • GDP.pdf

    4.0%

    Professional reasoning over PDF documents, all-pass rate

    Better than 17% of 171 scored models

  • AA-LCRv1.1

    52.0%

    Reasoning across long documents

    Better than 30% of 309 scored models

  • MLCR-AA

    1.7%

    Long-context reasoning over medical records

    Better than 3% of 121 scored models

Reasoning and knowledge

  • Humanity’s Last Exam

    19.6%

    Expert-level questions across academic fields

    Better than 47% of 323 scored models

  • GPQA Diamond

    78.2%

    Graduate-level science questions

    Better than 43% of 291 scored models

  • MMLU-Pro

    80.8%

    Multi-subject knowledge and reasoning

    Better than 53% of 132 scored models

  • IFBench

    69.0%

    Following precise output instructions

    Better than 69% of 215 scored models

Coding

  • Terminal-Benchv4.0

    0.0%

    Practical software engineering in a terminal

    Better than 0% of 171 scored models

  • SciCode

    34.0%

    Scientific computing in Python

    Better than 10% of 174 scored models

  • LiveCodeBench

    87.8%

    Recent competitive programming problems

    Better than 96% of 129 scored models

Math

  • AIME 2025

    93.4%

    Competition mathematics, 2025 problems

    Better than 93% of 118 scored models

Legacy evaluations

  • Terminal-Bench Hard

    23.5%

    Earlier terminal task suite

    Better than 44% of 212 scored models

  • τ²-Bench Telecom

    65.8%

    Tool use in a customer-support simulation

    Better than 50% of 214 scored models

  • AA-LCR (unversioned)

    52.0%

    Long-context reasoning, earlier grading

    Better than 30% of 309 scored models

Scores from Artificial Analysis, updated Oct 26, 2026. GPT OSS 120B was tested at high reasoning effort; compare scores measured at the same effort.Artificial Analysis

Live routing data

Performance and reliability

Recent averages across requests routed for this model. Your own speed depends on prompt length, reasoning effort and load.

Throughput
126.6 tok/s
Faster than 73% of 256 models
Time to first token
0.8 s
Quicker than 99% of 256 models
Prompt cache hits
7.1%
About $0.03 per 1M input tokens with caching
Request success (24h)
99.8%
Reliable

Cache hits are the share of input tokens recently served from a prompt cache across all requests. They are not a forecast of your own hit rate.

Hourly request success · last 72 hoursLowest hour 79.3%
  • Healthy (98% or more)
  • Degraded (90–98%)
  • Down (under 90%)
  • No requests

Developers

Use it through the API

Noumen’s API is OpenAI-compatible: any Chat Completions client works with a base URL and a key.

Read the quickstart
curl https://www.noumen.xyz/api/v1/chat/completions \
  -H "Authorization: Bearer $NOUMEN_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "openai/gpt-oss-120b",
    "messages": [{ "role": "user", "content": "Explain this stack trace." }]
  }'
import os
from openai import OpenAI

client = OpenAI(base_url="https://www.noumen.xyz/api/v1", api_key=os.environ["NOUMEN_API_KEY"])
reply = client.chat.completions.create(
    model="openai/gpt-oss-120b",
    messages=[{"role": "user", "content": "Explain this stack trace."}],
)
print(reply.choices[0].message.content)

OpenAI

Other OpenAI models, starting with the same family.