Blogs

We reverse-engineered Flash Attention 4

Modal

Asynchrony, fast approximate exponents, and 10x more efficient softmax.

Visit Site

We reverse-engineered Flash Attention 4
Best practices for serverless inferenceModal 8 Top Open-Source OCR Models Compared: A Complete GuideModal Google Cloud Run functions pricing: understanding costs and optimizationModal Fine-tuning a FLUX.1-dev style LoRAModal Azure Functions pricing: Consumption vs. Flex ConsumptionModal Glossary: LLM fine-tuning hyperparametersModal Docker Desktop 4.23: Updates to Docker Init, New Configuration Integrity Check, Quick SearchDocker Reverse proxy - MintlifyMintlify Frontend Proxy (Envoy)Opentelemetry Did You Know the Ordered List Element Has Start and Reversed Attributes?Css Tricks Which is faster: Gemini 3.5 Flash or Kimi K2.6 on CerebrasCerebras Adobe Flash - GlossaryDeveloper Mozilla GSK (Dreaming Big with Cerebras): Dr. Kim Branson - CerebrasCerebras Heart of dark mode: done, and still in progressTailscale How to Reverse CSS Custom CountersCss Tricks How to Create Microlearning Videos at ScaleSynthesia Gild Just One LilySmashingmagazine DeepSeek-V4.1-Flash on Fireworks: Astra-level DeepSWE at 1/15th the costFireworks Top 10 Static Site Generators (SSGs) 2021Netlify SnapKV: LLM Knows What You are Looking for Before GenerationCohere