Inco AI

Inference, reimagined for the agentic era.

[Inco AI @inco_ai]

Hope you enjoy our first release! (and more to come)

[Zhijian Liu @zhijianliu_]

DFlash 2 is here! Qwen3.8-27B at 70 tok/s on an M5 Max MacBook Pro.

⚡ Up to 4.6× the speed of autoregressive decoding, with the same output.

This is the next generation of DFlash, seeded at Z Lab and upgraded at Inco AI. Get one more accepted token on every pass, for free!

https://t.co/We0lwYPSBl

[SGLang @sgl_project]

Just pushed DFlash2 (@inco_ai) recipes to the Qwen3.8 27B cookbook⚡️

https://t.co/eAe1dFO5Sr

The community has been seeing great results with NVFP4 + DFlash2, and these recipes should be some very good starting points to play with.

More Qwen3.8 27B updates on the way 🫡

[CoreWeave @CoreWeave]

DFlash runs by default on our Kimi K2.7 Code endpoint, where we lead every provider in output speed on @ArtificialAnlys.

Congratulations to @inco_ai on DFlash 2. Thanks for the shoutout.

We look forward to even higher speeds ⚡

[Inco AI @inco_ai]

@CoreWeave @zhijianliu_ 💙

[Lucas Liebenwein @LLiebenwein]

We’ve been hard at work these past few months, and this is just the beginning 🚀

Today, we’re announcing DFlash 2, one of many innovations powering the inference stack at @inco_ai.

DFlash 2 delivers up to a 4.6× speedup over standard decoding, unlocking major performance gains for both frontier-scale models and smaller models running on your personal AI computer.

[Zhijian Liu @zhijianliu_]

DFlash 2 is here! Qwen3.8-27B at 70 tok/s on an M5 Max MacBook Pro.

⚡ Up to 4.6× the speed of autoregressive decoding, with the same output.

This is the next generation of DFlash, seeded at Z Lab and upgraded at Inco AI. Get one more accepted token on every pass, for free!

https://t.co/We0lwYPSBl

[Inco AI @inco_ai]

Hello, world!