Has there been any clear information from Meta over whether their new superintelligence initiative is going to still be open weights first? Or will they follow Google & OpenAI and keep frontier models closed & release less capable models open weights? Or are they entirely closed?
LLMS
-
H200 Hardware Achieves 10k Tokens Per Second Performance
By
–
it's an h200! i'm pretty sure you can get up to 10k tokens/second with some configurations
-
Leaked AI benchmarks: Real or fake? Community weighs in
By
–
Abrimos apuestas! ¿Los benchmarks filtrados la semana pasada serán reales o no?
-
AI Service Reliability Worth Five Percent Premium Cost
By
–
I think it’s like 5%, Totally worth it for the uptime, reliability, model switching.
-
Switching to OpenRouter for Direct Claude Access
By
–
I switch my provider to OpenRouter, and use Claude directly from there!
-
Major Partnership Announcement Supporting OpenAI Foundation Work
By
–
We’re proud to partner with @OpenAI and continue powering the foundation of their work.
-
Upgrading AI Coding Workflow: Switch to Claude Max
By
–
If you're switching away from Cursor, here's EXACTLY how to upgrade your AI coding workflow (I've tested everything — this setup is unbeatable): 1. Get Claude Max from Anthropic Anthropic currently subsidizes Claude, giving you thousands of dollars' worth of model usage for
-
Gemini 2.5 Pro performance: fast but unreliable, 2 correct out of 7
By
–
A few people have reported Gemini 2.5 Pro can do this as well. It can, and is much faster, but less reliable—in 7 attempts of my first prompt above, it returned: – 3 wrong answers (e.g. 2 shown below)
– 2 incomplete answers (omitting the cursive part)
– 2 fully correct answers -
Goodside seeks kindergarten path-following puzzles that fail in o3-pro
By
–
If you have good example of this you’d like me to try in o3-pro (that fails in other models) feel free to DM and I’ll give it a shot. Diagrams are harder in general; I’ve seen kindergarten-level path-following puzzles that it can’t do at all.
-
Sycophancy and Model Steerability: Control vs Instruction Following
By
–
I would argue there is a second problem, whose dimensions are unclear, which is whether sycophancy is part of steerability as well. A model that isn't sycophantic may not follow instructions?