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. 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.
openai/gpt-oss-120b
- Context
- 131k tokens
- Completion cap
- 131,072
- Tools
- Yes
- JSON
- Yes
- Released
- 2025-08-05
Call it from Geek Hub
Same OpenAI SDK. Change the base URL and the model id.
curl https://api.geekhub.mx/v1/chat/completions \
-H "Authorization: Bearer $GEEKHUB_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "openai/gpt-oss-120b",
"messages": [{"role": "user", "content": "Hola"}]
}'Get an API keyFrequently asked questions
- What is 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. 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. gpt-oss-120b runs on the Geek Hub API (OpenAI-compatible). Model id: openai/gpt-oss-120b.
- Is gpt-oss-120b free?
- No. Input is $0.0327 per 1M tokens and output is $0.1853 per 1M tokens on Geek Hub (markup included).
- What is the context length of gpt-oss-120b?
- gpt-oss-120b has a 131k tokens context window. It supports up to 131,072 completion tokens.
- Does gpt-oss-120b support tool calling and structured outputs?
- gpt-oss-120b accepts tools and tool_choice for function calling. It also supports structured outputs via a JSON schema in response_format.
- When was gpt-oss-120b released?
- gpt-oss-120b was released on 2025-08-05.
More models from OpenAI
GPT Image 1 (legacy)GPT Image 1 moved to tokenized pricing ($10/$40 per 1M tokens). We show approximate price per 1024×1024 high-quality image. A new 'image-tokens' kind is coming in a future sprint.
GPT-4.1 mini (legacy)1M context at low price. Replaceable by GPT-5.6 Luna or Gemini 2.5 Flash-Lite.
GPT-4o mini TTSLatest generation: prompt-controlled voice (instructions for tone, emotion, pace). 7 voices.
GPT-5 (legacy)Original GPT-5. Kept for backward compat; recommend migrating to 5.1 or 5.4-mini depending on use case.
GPT-5.1Refinement of GPT-5 at the same price. Better instruction following and tool handling.
GPT-5.4 miniEconomic variant of 5.4. Sub-dollar per 1M input tokens, good balance for chatbots.