What’s Cooking

CerebrasBlog2w

Why AI Assistants Are Slow—and How to Make Them Faster

ModalBlog2w

Hitting a billion tokens per minute on one GPU by combining a query planner and an inference engine

Fireworks AIBlog3w

Introducing Ember-1

ModalBlog3w

How to serve trillions of tokens for trillion-parameter coding agents

NVIDIABlog3w

Simplifying Model Serving Across Multiple GPUs with NVIDIA TensorRT Multi-Device Integration in NVIDIA Dynamo-Triton

Fireworks AIBlog3w

The frontier isn’t a model. It’s a router.

AirbnbBlog3w

The guest journey, updated in real time: extending Airbnb’s sequence recommender with Chronon

VercelBlog3w

Open-weight models take 56% of token volume, Astra doubles Fable 5.1 spend

GoogleBlog4w

Agent Anomaly Detection, now in Private Preview on the Gemini Enterprise Agent Platform- Google Developers Blog

LambdaBlog4w

MLPerf Inference v6.1: pioneering agent, VLM benchmarks

AppleBlog4w

DACA-GRPO: Denoising-Aware Credit Assignment for Reinforcement Learning in Diffusion Language Models

Fireworks AIBlog4w

DeepSeek-V4.1-Flash on Fireworks: Astra-level DeepSWE at 1/15th the cost

BasetenBlog4w

DeepSeek-V4.1-Flash: more efficient prefill for coding agents

LambdaBlog4w

OpenResearcher: a reproducible and scalable pipeline for training deep research agents

BasetenBlog4w

How Baseten makes pyannote’s diarization models 9.6x faster

GitLabBlog8 Sept

GPT-6 Astra on GitLab: Faster runs, fewer tokens used

ElasticBlog4 Sept

Why your AI bill tripled while token prices fell 75%

DecagonBlog4 Sept

How we made one of our largest inference workloads 4.7× more GPU-efficient

InstacartBlog3 Sept

Agentic Machine Learning Modeling at Instacart

NVIDIABlog2 Sept

Co-Designing AI Models Using Speculative Decoding for Faster LLM Inference

BasetenBlog2 Sept

The efficient frontier of LLM inference

DeepMindBlog1 Sept

Introducing agentic video understanding with Gemini

You’re caught up

That’s every post that month in this feed.