Announcing our Series F. Learn more
Blog

Inference engineering

AllResearchModel performanceAI engineeringInfrastructureNewsCommunityAI modelsProductFoundations
Model performance

How we built the fastest GLM 5 API

Tri Dao
2 others
Research

Repeated KV cache for long-running agents

Charles O'Neill
AI engineering

Outperforming frontier models on emergency medicine documentation

Harry Partridge
2 others
Research

Introducing RadixMLP: Intra-batch deduplication for causal transformers

Michael Feil
Model performance

Four Bits: 4-Bit Quantization for Inference Optimization in FLUX.2 [dev]

Ali Taha
News

Announcing the acquihire of Inferless by Baseten

Nilesh Agarwal
1 other
Position

The Michael Scott Paper Company of AI

Charles O'Neill
AI models

MiniMax M2.5: Intelligence too cheap to meter, RL process rewards, real-world productivity

Alex Ker
Research

Distillation without the dark

Max Kirkby
2 others
1...567...25