LLM Perks

gpt-oss-120b

OpenAI · All GPT models
Free right now Frontier
ReasoningCodingTool useOpen weights

gpt-oss-120b is a language model with 117B parameters from OpenAI. It is currently free at 3 providers with up to 131K tokens of context. The model supports step-by-step reasoning, is suited to coding, supports tool calling. Its weights are open, so it can also run locally.

Developer
OpenAI
Parameters
117B / 5.1B active
Context
131K tokens
Type
Text
License
Open weights
Free since
September 19, 2026

Where to use gpt-oss-120b for free

Each row is a separate free endpoint with its own limits.

ProviderContextLimitCard
Groq
openai/gpt-oss-120b
131K 14,400 req/day No Open
Cloudflare Workers AI
@cf/openai/gpt-oss-120b
128K Workers Free: 10,000 Neurons/day shared across eligible models — Open
UnoRouter
gpt-oss-120b:free
131K No-card free models use shared capacity and per-model limits; availability can change. No Open

Quick start

curl https://api.groq.com/openai/v1/chat/completions \
  -H "Authorization: Bearer $GROQ_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "openai/gpt-oss-120b",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'

Example for Groq. Any OpenAI-compatible client works: Cursor, Cline, OpenCode, Open WebUI.

Benchmarks

Intelligence Index
11.6
GPQA Diamond
78.2%
Humanity's Last Exam
19.6%
Terminal-Bench Hard
23.5%
τ²-Bench
65.8%
Long context (LCR)
52.0%

Source: Artificial Analysis · gpt-oss-120b (High)

Similar free models

FAQ

Is gpt-oss-120b free?

Yes. As of October 24, 2026, gpt-oss-120b is free at Groq, Cloudflare Workers AI, UnoRouter. Each provider has its own limits — see the table above.

What is the context window of gpt-oss-120b?

The largest context among free routes is 131,072 tokens. Some providers cap it lower.

How do I call gpt-oss-120b via API?

Use any OpenAI-compatible client with base URL https://api.groq.com/openai/v1, model ID openai/gpt-oss-120b, and a Groq key.

How good is gpt-oss-120b?

Artificial Analysis Intelligence Index: 11.6, GPQA Diamond: 78%. For comparison, the strongest free model in the catalog right now is DeepSeek V4.1 Flash.

Is gpt-oss-120b open-weight?

Yes, the weights are published — you can download and run it locally, for example with Ollama or LM Studio.