$SPCE to the moon 🚀🌕💎

Why are airport immigration not ALWAYS fully staffed? What do you mean you have hundreds of thousands of passengers a day in traffic but you can't staff 10 people full time to speed up the queue?

Anthropic now valued at $965 billion with $47 billion run-rate revenue. They just released Claude Opus 4.8, a modest bump on 4.7 in benchmarks but still put them firmly as the top LLM lab. GPT-5.6 coming soon to retake the crown?

Is it just me or did codex/gpt-5.5 got really dumb recently? Writing huge messy diffs, tripping up on shell quoting every other bash call and compact loses context faster than before. Time for a reset @thsottiaux?

Man used to go to war, now they compare 20 months old model against 38 months old model.

There has not been an update

It's 2026, how tf are most models/inference providers still not supporting reasoning real structured output?? This is already implemented by DeepSeek R1 at the start of 2025 - ancient times in the AI space.

Good overview on full-duplex models (omni-to-speech models that can perceive and output at the same time). Progress in this space has been slow relative to other areas in AI primarily due to the lack of training data. Hope @thinkymachines new research can accelerate it!

Man used to fight gorillas for food, now they make dumb posts about a GPT wrapper

Superintelligence will be achieved by whoever takes the bitter lesson the most serious: A single all-to-all MLP with no human shortcuts, RL'ed in a equally bitter-lessoned world model.

Longest codex /goal run yet at 22 hours! I am optimizing Qwen3 Omni on SGLang to be able to serve 30 concurrent real time voice chats on an AMD MI300X. Starting with one active chat not even hitting real time, it's almost reaching the concurrency goal after almost a day of optimization and profiling. I thought it was cool how people were able to get agents to run automonously for a long time, turns out this is actually boring and unproductive. Still a long way to go for LLMs to get so good they can think for a few minutes and oneshot the solution. Make no mistake, the frontier models are very much AGI, but I can't wait for superintelligence to come fast enough.

bro disappeared like he never existed

AMD is fumbling AI so hard right now. They have equal if not better chips than Nvidia, but NOTHING works on it. Most models fail to run apart from using the most naive fallback paths. Flash attention 3/4 is not implemented. They could win this so easy if they just hire a team of kernel wizards to get most of the models on vLLM/SGLang to work and be optimized on AMD GPUs.

OpenAi/Anthropic/literally every lab: introducing the world's best model xAI: we go to the office 7 days a week!!

i should never be sharing this but f*ck it seedance 2.0 claude code tiktok is THE best combo for AI videos i cracked the formula for generating videos that look hyper realistic & make your audience feel like "holy shit this person gets me" i'm finally sharing my FULL system with you.. here's what you're getting: - my prompting method for realistic voices (works every time) - my realistic human movements & breathing claude skill (this is key for realistic videos) - my exact method on how to make infinite length videos that maintain consistency - how to get AI tools for dirt cheap (90% off) RT reply 'UGC' and i'll send you the step-by-step system (must follow so i can dm)

$20,366 in 2 Weeks spamming tiktok slideshows - How to target USA - How to bypass tiktok's duplicate detection - How to PRINT with slideshows.. Retweet & Comment "Slideshow" (must be following) and I'll send you the link to the full guide.