To those in the replies who say 'but opus 4.8 is weaker so without fallback, the score would be even higher': this is not necessarily true because of how any benchmark works – which is an average of queries – and what is called 'the'
LLMS
-
SambaNova congratulates MiniMax on M3 open-weight model launch
By
–
Congrats to our partners at @MiniMax_AI on the launch of MiniMax M3. Open-weight models continue to push the ecosystem forward, and we're excited to bring M3 to RDUs down the road. Looking forward to following what's built with it.
-
Data work key to model improvement: evals, rubrics, error analysis
By
–
obviously, but u are missing the point i'm saying most of what makes models better is DATA WORK: evals, rubrics, error analysis, and so on Fable can do this, Mythos will do this, future models will do this
-
Kimi K2.7 Code Proves Less Thinking Improves Coding Performance
By
–
The less your model thinks, the better it codes. Kimi K2.7 Code proved it. The entire reasoning model space has been moving in one direction. More thinking tokens, longer chains, bigger reasoning budgets. Kimi just challenged that. K2.7 Code scores higher than K2.6 on every
-
Kimi K2.7 Code and MiniMax M3 available on Poe
By
–
Kimi K2.7 Code and MiniMax M3 are now available on Poe. Two cutting-edge open-weight models designed for serious coding and agentic tasks. Kimi K2.7 Code is Moonshot AI's most powerful coding model to date, with multi-step tool calling more
-
Huge leverage in improving data and xAI’s plan to fix pretraining typos
By
–
An underrated part of this discussion is that (a) there's huge leverage in improving data, and (b) there's no way Anthropic could safeguard this xAI could instruct Fable to look through EVERY row of pretraining data and fix any typos and errors. this probably the single
-
Olivier Rimmel congratulates Mistral and calls for three AI providers in France
By
–
The French AI #Mistral is very advanced and very performant, not at the level of Opus, even less of Fable, but it's still very good and it's improving all the time. I want three providers like that in France. President @EmmanuelMacron, make sure there are
-
Working on fixes for prompt and future models as stopgap
By
–
We're working on a few potential fixes for this in the prompt and in future models. This is more of a stopgap until those land
-
Tip to make Fable write clearly in agentic conversations
By
–
Fable seems superhuman in managing long agentic conversations, sometimes to the point where I can't follow what it tells me. This prompt snippet has been the best solution I've found to make it write clearly and abandon all jargon:
-
Boris, creator of Claude Code, only writes loops now
By
–
EL CREADOR DE CLAUDE CODE YA NO ESCRIBE PROMPTS
— Nico (@nicos_ai) 12 juin 2026
Boris, la persona que construyó la herramienta, soltó esto en una entrevista:
"Ya no escribo prompts. Escribo loops, y los loops hacen el trabajo. Mi trabajo es escribir loops."
Mientras la mayoría de devs sigue puliendo el… https://t.co/yK038QMxcc pic.twitter.com/fJ9336DkTVTHE CREATOR OF CLAUDE CODE NO LONGER WRITES PROMPTS Boris, the person who built the tool, said this in an interview: "I don't write prompts anymore. I write loops, and the loops do the work. My job is to write loops." While most