Kimi K3 is here. Try it now

Michael Feil

Model Performance Engineer

Model performance

Making Kimi K3 tokenization 18x faster for million-token agentic workloads

Michael Feil
Model performance

Sub-3 millisecond named entity recognition (NER) inference

Michael Feil
1 other
Model performance

How we built the fastest Kimi K2.5 on Artificial Analysis

Tri Dao
3 others
Model performance

2x faster inference with KV cache-aware routing

Abu Qader
2 others
Model performance

Run Qwen3 Embedding on NVIDIA Blackwell GPUs

Michael Feil
1 other
AI engineering

Your client code matters: 12x higher embedding throughput with Python and Rust

Michael Feil
Model performance

Day zero benchmarks for Qwen 3 with SGLang on Baseten

Yineng Zhang
2 others
Model performance

How we built BEI: high-throughput embedding, reranker, and classifier inference

Michael Feil
1 other
News

Introducing Baseten Embeddings Inference: The fastest embeddings solution available

Michael Feil
1 other

Explore Baseten today

Start deployingTalk to an engineer