RedHatAI/gpt-oss-20b

TEXT GENERATIONConcurrency Cost:1Model Size:20BQuant:FP8Ctx Length:32kPublished:Sep 4, 2025License:apache-2.0Architecture:Transformer0.0K Open Weights Cold

RedHatAI/gpt-oss-20b is a 21 billion parameter open-weight language model from OpenAI, designed for powerful reasoning, agentic tasks, and versatile developer use cases. It features configurable reasoning effort, full chain-of-thought access, and agentic capabilities including function calling, web browsing, and Python code execution. This model is optimized for lower latency and specialized applications, running efficiently within 16GB of memory due to MXFP4 quantization.

Loading preview...

RedHatAI/gpt-oss-20b: An Open-Weight Model for Reasoning and Agentic Tasks

RedHatAI/gpt-oss-20b is a 21 billion parameter open-weight model from OpenAI, part of the gpt-oss series, designed for robust reasoning and agentic capabilities. It is specifically tailored for lower latency and specialized use cases, capable of running within 16GB of memory thanks to MXFP4 quantization of its MoE weights.

Key Capabilities

  • Permissive Apache 2.0 License: Allows for free experimentation, customization, and commercial deployment without copyleft restrictions.
  • Configurable Reasoning Effort: Users can adjust the model's reasoning depth (low, medium, high) to balance speed and detail for specific tasks.
  • Full Chain-of-Thought Access: Provides complete visibility into the model's reasoning process, aiding debugging and increasing trust in outputs.
  • Fine-tunable: The model can be fully customized for specific use cases, with this smaller version being fine-tunable even on consumer hardware.
  • Agentic Functionality: Natively supports function calling, web browsing, Python code execution, and structured outputs.

Good For

  • Developer Use Cases: Its agentic capabilities and fine-tunability make it suitable for integrating into various development workflows.
  • Specialized Applications: Ideal for scenarios requiring lower latency and efficient memory usage.
  • Experimentation and Customization: The Apache 2.0 license and fine-tuning support encourage broad application development.
  • Debugging and Trust: Full chain-of-thought access is beneficial for understanding and validating model decisions.

Finetunes
3 models