Now it's time to test the model to draw clear conclusions not just from benchmarks, but from the real vibes of working with it. And also wait for external evaluations on the rest of the benchmarks from those who go on testing via the API. We'll keep you posted
LLMS
-
GPT 5.5 Frontend Design Issues Persist Despite Updates
By
–
En mi carta a Santa Claus puse que ojalá GPT 5.5 arreglara por fin el problema de diseño de frontends, pero claro… en Abril no viene Santa Claus.
— Carlos Santana (@DotCSV) 23 avril 2026
Noto ligeras mejoras pero ahí están, las cajas azules rodeándolo todo.https://t.co/BMoRdNcIxDIn my letter to Santa Claus, I wrote that I hoped GPT 5.5 would finally fix the frontend design problem, but of course… Santa Claus doesn't come in April. I notice slight improvements but there they are, the blue boxes surrounding everything.
-
ARC-AGI 5.5 outperforms Gemini 3.1 Pro at reasoning
By
–
In ARC-AGI 5.5, it is positioned at the cost frontier of Gemini 3.1 Pro but achieving a better score in its highest reasoning levels.
-
GPT-5.5 Live Launch Announced by OpenAI
By
–
GPT-5.5 Live with Every https://
x.com/i/broadcasts/1
AxRnaLmZYPxl
… -
OpenAI’s Rapid GPT Release Pace Raises Marketing Concerns
By
–
I'm sure that GPT 5.5 by vibes will be an excellent model, just like its predecessor was. And hey, OpenAI's pace of progress is excellent if we think about the fact that the previous one came out a month and a half ago! BUT FOR GOODNESS' SAKE OPENAI STOP WITH THE OVER-THE-TOP
-
GPT-5.5 executes plans without hesitation
By
–
many models write a great plan and then get scared of it.@OpenAI's new GPT-5.5 just does the thing.
— Dan Shipper 📧 (@danshipper) 23 avril 2026
here's what changed 👇 pic.twitter.com/Z79GUK7Opdmany models write a great plan and then get scared of it. @OpenAI
's new GPT-5.5 just does the thing. here's what changed -
New Model Costs 20% More Than GPT 5.4 Despite Token Efficiency
By
–
Because the price of the model doubles (!!) its predecessor, buuuut this is something that's mitigated by the more efficient use of tokens But in combination, as a result, it leaves us with a model that's 20% more expensive than GPT 5.4 xhigh when it comes to evaluating
-
Codex and GPT-5.5: Combining Models for Shipping Work
By
–
Codex + GPT-5.5 is an incredible combo.
— Romain Huet (@romainhuet) 23 avril 2026
A great model needs a great space to work. Codex gives GPT-5.5 the context, tools, artifacts, browser, and review loop to turn reasoning into work you can ship. https://t.co/9sAnYFBDdlCodex + GPT-5.5 is an incredible combo. A great model needs a great space to work. Codex gives GPT-5.5 the context, tools, artifacts, browser, and review loop to turn reasoning into work you can ship.
-
User claims AI model ‘Claude’ was intentionally degraded
By
–
I can't believe we were right Claude was dumbified on March 4, just when we noticed!
-
GPT 5.5 Base64 Encoding Security Capabilities
By
–
Il doit sûrement être bon en gestion de hack en base64 le #gpt5.5