gpt-oss-120B
GPT-OSS-120B is a 120 billion parameter open-weight language model released by OpenAI under Apache 2.0 licensing, designed for reasoning tasks and optimized to run on high-end consumer hardware such as H100 GPUs. Recent research has shown the model struggles with instruction-following during reasoning processes and exhibits cascading failures in IT automation tasks compared to frontier models, while it is now available for fine-tuning and commercial deployment through infrastructure providers like Together AI.
Updated 3 August 2026
Specifications
No specifications recorded yet.
Latest developments
IBM and UC Berkeley Diagnose Why Enterprise Agents Fail Using IT-Bench and MAST
Hugging Face Blog · 5 months ago ·
41
Large Reasoning Models Fail to Follow Instructions During Reasoning: A Benchmark Study
Together AI · 9 months ago ·
37
Transform OpenAI gpt-oss Models into Domain Experts with Together AI Fine-Tuning
Together AI · 11 months ago ·
20
From GPT-2 to gpt-oss: Analyzing the Architectural Advances
Ahead of AI · 11 months ago ·
30
Announcing the Availability of OpenAI's Open Models on Together AI
Together AI · 11 months ago ·
49
Introducing gpt-oss
OpenAI Blog · 11 months ago ·
46
gpt-oss-120b & gpt-oss-20b Model Card
OpenAI Blog · 11 months ago ·
26
Q1 2026
Q4 2025
Q3 2025
OpenAI releases two open-weight models gpt-oss-120b and gpt-oss-20b under Apache 2.0 license Open source release
OpenAI Releases GPT-OSS Open-Weight Reasoning Models Under Apache 2.0 License Open source release
Relationships
Products & technology
- OpenAI develops this model · 5 sources
- Together AI deploys this model · 1 source
- Together AI integrated with this model · 1 source