Tag
Together AI has open-sourced a complete recipe, allowing you to fine-tune your own Jev classification model for just $17, and has released a new model based on Qwen3.5 4B.
The article announces the launch of together/Tev1-4B-experimental, a Jev-like classification model fine-tuned from Qwen3.5 4B, and provides a guide on training your own version for approximately $17.
Profiles Dan Fu, a key contributor to high-performance kernels like FlashAttention, Hyena, Monarch Mixer, and ThunderKittens, now a distinguished researcher at Together AI whose work is used in ChatGPT, Claude, and Gemini.
Dan Fu announces a live chat with Olive Song at aiDotEngineer about MiniMax 3, covering its training decisions.
Together AI open-sources OSCAR, an attention-aware 2-bit KV cache quantization system that enables efficient long-context LLM serving by redistributing quantization error according to attention importance.