All AI models
Open-weightVerified August 11, 2026

OpenAI

gpt-oss-120b

gpt-oss-120b is the high-capacity choice in OpenAI's open-weight line when you want a permissive license, strong text reasoning, and control over where model data runs. It makes sense for high-memory GPU infrastructure or a managed host, not as a casual local model download.

License
Apache 2.0
Context
128K tokens
Modalities
Text
Parameters
117B total, 5.1B active per token

What the license means

OpenAI releases gpt-oss weights under Apache 2.0, subject to the gpt-oss usage policy. OpenAI distinguishes the open weights from its hosted API and ChatGPT products.

Open-weight, open source, and no-cost API access mean different things. Always review the linked terms and your own data, hosting, and compliance requirements before shipping.

API and operating cost

Input / 1M
No OpenAI API price
Output / 1M
No OpenAI API price

gpt-oss is not served through the OpenAI API or ChatGPT. Your cost is the infrastructure or third-party hosting price you choose.

Check current provider price

Where it fits

  • Private-cloud reasoning services
  • High-capacity text agents
  • Custom policy and tool workflows

Where it does not fit

  • Small-memory local machines
  • Teams expecting OpenAI API billing or ChatGPT access

Local deployment reality

OpenAI states that the native MXFP4 120B release can run within 80GB of memory. It is a realistic single high-memory GPU or hosted-inference option, not a typical consumer laptop model.

vLLMOllamallama.cppTransformers

Practical setup notes

  • Use the Harmony prompt format and a supported renderer for reliable behavior.
  • Treat hardware, storage, and operations as part of the total cost, not an afterthought.

Capability notes

Reasoning
Configurable reasoning effort through supported runtimes
Tools and output
Tool use, function calling, structured output through common runtimes

Limitations to plan around

  • Text-only model.
  • Not available in OpenAI API or ChatGPT.

Continue comparing

Related model decisions

See all reviewed models
Open-weight

gpt-oss-20b

Local text reasoning

Open source

Qwen3.6

Local coding assistance

Open source

Mistral Small 4

Multimodal document analysis

Editorial note: This is a dated decision page, not a benchmark leaderboard or a substitute for legal, security, or infrastructure review. Model and pricing facts can change after August 11, 2026; use the linked primary sources before committing budget or customer data.