Blog
What the lab publishes.
Writing from the team, and release notes for everything we ship. Newest first.
- Nov 2026BlogFull speed, full precision: Up to 12.8× higher agent interactivity than SGLang on Qwen3.8-Flash-NextRead
- Nov 2026BlogQwen3.8-27B: Outperforming SGLang and vLLM Across the BoardRead
- Nov 2026BlogUp to 95% Higher Qwen3.6 Throughput Than vLLMRead
- Nov 2026Blog4.94× Higher Sustained Throughput Than vLLM on GLM-5.3Read
- Oct 2026ChangelogBackboard Update: fal.ai Tools + Product UpdatesRead
- Sep 2026ChangelogReal-time Voice-to-Voice Is HereRead
- Sep 2026ChangelogVideo Generation, Plus Image API ImprovementsRead
- Sep 2026BlogThe Era of Static Software Is EndingRead
- Aug 2026BlogThe Hidden Parts of the AI StackRead
- Aug 2026BlogThe AI Industry Has a Benchmark ProblemRead