Increase inference performance by up to 15x without sacrificing responsiveness. DFlash, an open source lightweight block diffusion model designed for speculative decoding, delivers up to 15x higher throughput on NVIDIA Blackwell while maintaining the same user interactivity
AI
-
Codex is my go-to backup for these moments
By
–
haha Codex has honestly become my go-to backup for exactly these moments.
-
Thousands of agents in loops overnight is not everyone’s reality yet
By
–
"this is just how engineering is done now" is doing a lot of work haha, thousands of agents in loops overnight is his reality, not exactly everyone's yet.
-
Happy with Codex over Claude Code until Fable returns
By
–
haha honestly been surprisingly happy with Codex over Claude Code lately, at least until Fable's back.
-
Latitude groups repeated failures with frequency and reason
By
–
Been waiting for someone in agent observability to do this properly. Latitude doesn't just hand you traces.
— Chubby♨️ (@kimmonismus) 23 juin 2026
It collapses repeated failures into one issue with frequency and reason, and lets you search production conversations in plain English.
Open source, MIT licensed,… https://t.co/8ehTAM7rzqI was waiting for someone in agent observability to do this correctly. Latitude doesn't just give you traces. It groups repeated failures into a single problem with frequency and reason, and allows you to search production conversations.
-
Mythos-level AI risks and open release preparation concerns
By
–
All Mythos-level models are likely to invite similar risks. Those risks will only be greater with the release of open Mythos-class AI coming in the next 6-12ish months (assuming China allows it) The lack of clarity over what risks concern the government may be slowing preparation
-
@testingcatalog — 2026-06-23
By
–
Mistral AI a lancé OCR 4 > Taux de réussite moyens de 72 %, accompagnés du meilleur score global sur OlmOCRBench (85,20).
> En plus du texte extrait, OCR 4 renvoie des boîtes englobantes, une classification de blocs typés, et des scores de confiance en ligne.
> OCR 4 est un -
AI in 2040: Nearly Optimal Stack, Massive Current Inefficiency
By
–
AI in 2040 will not be built on the stack we use today. It will be much closer to optimal. The current stack exhibits 3-4 orders of magnitude data inefficiency and 4-5 orders of magnitude compute inefficiency. Nearly optimal AI is what
-
Learn Vercel’s new Eve agentic framework with hands-on labs
By
–
Learn to use the new eve agentic framework from Vercel.
— DAIR.AI (@dair_ai) 23 juin 2026
Go try out the hands-on labs now. https://t.co/mC5nsSUNOeLearn to use the new eve agentic framework from Vercel. Go try out the hands-on labs now.