Hosted by @TogetherCompute, you can try gpt-oss-120b at and across all platforms. This model is also available via our new API. (2/2)

OpenAI's new gpt-oss is now on Poe! gpt-oss-120b is a state-of-the-art open-source model that delivers exceptional performance in complex reasoning, multi-step problem-solving, and analysis. (1/2)

New guide: OpenAI gpt-oss Quickstart! It includes: ◆ Use cases ◆ Best practices ◆ Managing context and reasoning effort ◆ Model architecture and limitations

🆕 See how gpt-oss-120b does against o4-mini in our latest video! We go over 5 simple tests across coding, reasoning, summarization, and web search to help give devs a feel of the model.

Opencode, the popular AI coding agent, now supports Together AI models!

🤖 𝐂𝐨𝐠𝐢𝐭𝐨 𝐯𝟐 from @DeepCogito just hit Together AI and these open-source reasoning models are approaching frontier performance 🤯 Four models spanning 70B dense, 109B MoE, 405B dense, and 671B MoE - designed to think smarter instead of just thinking longer.

🔥 FLUX.1 Krea [dev] just landed on Together AI Finally: text-to-image that doesn't look obviously AI-generated. This "opinionated" model from @bfl_ml x @krea_ai beats the oversaturated aesthetic problem that's been killing photorealism. 👇

🚨 New Talk for ML Practitioners! LLM-as-a-Judge is changing how we evaluate models - fast, consistent, automated. We’ll demo how to score SoTA open models (Kimi, Qwen, GLM), share prompting tricks, and unpack eval best practices. 🔗 + Q&A with Together AI team!👇

🧪 New Notebook Drop: Evaluating LLMs for harmful outputs! Which models are actually safe for prod? We built an LLM-as-a-Judge pipeline using the Together Evals API to compare models on harmfulness. Code👇

Run Qwen 3 Coder directly on the @huggingface model page – powered by Together AI!