NewsThe building blocks of the open frontier
This week we’re expanding our inference platform with three launches alongside our fresh new website.
Amir Haghighat
Please rotate your device
To continue exploring this website, please rotate your device to a vertical orientation
Blog
Inference engineering
NewsAmir Haghighat
ProductProductIntroducing Serverless Loops: frontier RL training for everyone
Kimbrian Canavan, Raymond Cano, Ardin Lo
Product
ProductProductBuilding the Agent Cloud: Introducing Baseten Sandboxes
Paul Sinai, Nicolas Lecomte, Sai Maddali, Rachel Rapp
- Model performance
How to make flash models even faster: optimizing DeepSeek V4.1 Flash 1017
Alex Korte, Alex Ker, Philip Kiely, Joyjit Daw - Model performance
The fastest delta weight syncs in the world
Jerry Hong, Paras Stefanopoulos, Ardin Lo - Model performance
The fastest, lowest-cost batch transcription with Qwen3-ASR
Hasan Khan, Ansel Erol - Model performance
What makes a high-quality NVFP4 quantization?
Chloe Florit - Model performance
Agentic inference optimization: 50-90% faster engines
Shawn Rushefsky - Model performance
How Baseten makes pyannote’s diarization models 9.6x faster
Matte Lim, Ansel Erol - Model performance
Optimizing delta weight syncs for managed RL rollouts
Paras Stefanopoulos, Ardin Lo - Model performance
Baseten leads Coval’s voice AI benchmark
Ansel Erol, Alina Weinstein - Model performance
The efficient frontier of LLM inference
Philip Kiely
Bring open models to production today.
