T. Rebentisch

[T. Rebentisch @RebentischT]

@AiBattle_ its official: the better model can easily be determined by which one can create the better minecraft temple!

[T. Rebentisch @RebentischT]

@scaling01 lmao 4.7 already better than 5.5

[T. Rebentisch @RebentischT]

@JesusGodAndKing @zkespresso @emanueledpt @elder_plinius you really think this? XD the "exploit" was just asking the model to fix a codebase without specifying if its security related or not. EVERY frontier model of the last 6 months can do that.

[T. Rebentisch @RebentischT]

@yoxics She Zhang et al.

[T. Rebentisch @RebentischT]

@bridgemindai Maybe consider your benchmark sucks lmao

[T. Rebentisch @RebentischT]

@ChatGPTapp frontend recursive ASI has been achieved

[T. Rebentisch @RebentischT]

@bridgemindai wym "for a fraction" XD its at least as pricy as GPT 5.5 which is 5 in 30 out. I mean yeah 1/2 is considered a fraction so "technically correct"

[OpenAI @OpenAI]

Today, we share a breakthrough on the planar unit distance problem, a famous open question first posed by Paul Erdős in 1946.

For nearly 80 years, mathematicians believed the best possible solutions looked roughly like square grids.

An OpenAI model has now disproved that belief, discovering an entirely new family of constructions that performs better.

This marks the first time AI has autonomously solved a prominent open problem central to a field of mathematics.

[T. Rebentisch @RebentischT]

@bridgemindai who cares about (proven to be poorly designed) benchmarks? lmao Gemini 3 had amazing benchmarks for its release date and it was unusable for agentic coding; only thing that matters is performance on real tasks

[Marcel @realmarcel1]

Presque 900000 signatures pour l’initiative citoyenne européenne pour la suspension totale de l’accord d’association UE-Israël
Il en faut un million ⬇️
https://t.co/EqnS9iRBkj