AI Lab · by Youssef Mahmoud · 2026-10-20
63 sparks · 16 comments
Kabir Khan · 8 likes
Tried it on my own dataset, results were mixed
Olena Shevchenko · 10 likes
Yeah same here
Lucía García · 10 likes
Ha, my wallet disagrees
Mango Coder · 2 likes
The cost per token matters more than the leaderboard imo
Chinedu Adeyemi · 0 likes
lol this made me laugh
Paper Runner · 2 likes
Prompt sensitivity is still the biggest pain in production
Jan Nowak · 0 likes
Fair point
Paolo Garcia · 2 likes
Evals are getting saturated, we need harder ones
Bilal Raza · 13 likes
Fair point
Sara Ahmed · 0 likes
True, learned it the hard way
Lukas Huber · 0 likes
Context window claims are always optimistic, test it yourself
Quiet Mango · 5 likes
True, learned it the hard way
Nima Ahmadi · 0 likes
My GPU just cried reading this
Rizky Pratama · 5 likes
This ^
Grace Parker · 0 likes
Open weights or it didnt happen
Hana Choi · 5 likes
Well said