Happy Horse 1.0 is now on ModelsLab

Try Now
Skip to main content
Available now on ModelsLab · Language Model

OpenAI: Gpt-oss-120bOpenAI gpt-oss-120b Power

Try OpenAI: Gpt-oss-120bAPI Documentation

Deploy gpt-oss-120b Efficiently

MoE Architecture

117B Params 5.1B Active

Activates 5.1B parameters per token in 36-layer MoE with 128 experts.

Single GPU Fit

Runs on H100

Fits in 80GB VRAM via MXFP4 quantization for efficient production inference.

Agentic Reasoning

Tool Use Native

Supports function calling, browsing, code execution with configurable effort levels.

Examples

See what OpenAI: Gpt-oss-120b can create

Copy any prompt below and try it yourself in the playground.

Code Optimization

“Analyze this Python function for efficiency: def fibonacci(n): if n <= 1: return n else: return fibonacci(n-1) + fibonacci(n-2). Suggest memoized version with benchmarks.”

Market Trend Report

“Summarize quarterly sales data trends from CSV: Q1:100k, Q2:120k, Q3:110k, Q4:150k. Forecast Q1 next year using linear regression.”

Tech Architecture

“Design scalable microservices for e-commerce: user auth, inventory, payments. Include Docker, Kubernetes, API gateways.”

Algorithm Explanation

“Explain A* pathfinding algorithm step-by-step with pseudocode. Compare to Dijkstra on grid graph example.”

For Developers

A few lines of code.
gpt-oss-120b. One API call.

ModelsLab handles the infrastructure: fast inference, auto-scaling, and a developer-friendly API. No GPU management needed.

  • Serverless: scales to zero, scales to millions
  • Pay per token, no minimums
  • Python and JavaScript SDKs, plus REST API
API Documentation 
import requests


response = requests.post(
    "https://modelslab.com/api/v7/llm/chat/completions",
    json={
  "key": "YOUR_API_KEY",
  "prompt": "",
  "model_id": ""
}
)
print(response.json())

FAQ

Common questions about OpenAI: Gpt-oss-120b

Read the docs

OpenAI: gpt-oss-120b is 117B-parameter MoE LLM with 5.1B active params per token. Matches o4-mini on reasoning benchmarks. Runs on single H100 GPU.

Access via LLM endpoint with Apache 2.0 license. Supports 131k context, reasoning tokens. Configure low/medium/high effort.

Yes, fine-tune on H100 node. Customize for specific tasks like coding or agents. Apache 2.0 allows commercial use.

Open-weight alternative to o3/o4-mini with near-parity performance. Strong on agentic tasks, tool use. No copyleft restrictions.

Single 80GB GPU like H100 or MI300X. Uses MXFP4 for MoE weights. 131k context window supported.

Text-only input/output. Full chain-of-thought access. Native function calling, browsing, Python execution.

Ready to create?

Start generating with OpenAI: Gpt-oss-120b on ModelsLab.

Try OpenAI: Gpt-oss-120bAPI Documentation