Four months have passed since the initial rollout, and the 5.1 iteration brought subtle refinements rather than a complete overhaul. The team skipped the playful codename this time—unlike previous naming conventions—because this feels different. We're witnessing a genuine leap forward here.
Hitting 82% on SWE Bench isn't just impressive; the trajectory suggests there's still room to climb. No plateau in sight yet.
This page may contain third-party content, which is provided for information purposes only (not representations/warranties) and should not be considered as an endorsement of its views by Gate, nor as financial or professional advice. See Disclaimer for details.
12 Likes
Reward
12
7
Repost
Share
Comment
0/400
MetaReckt
· 12-14 11:10
82% directly took off, this round was not a fake play
View OriginalReply0
GasGrillMaster
· 12-13 18:32
82%? That's a bit suffocating, feels like it can go even higher.
View OriginalReply0
PonziWhisperer
· 12-11 15:59
82% and still going higher? This update feels like it has something really substantial
View OriginalReply0
ContractHunter
· 12-11 15:58
82%—that's some real potential, especially since it hasn't peaked yet. Now that's what you call a promising stock.
View OriginalReply0
token_therapist
· 12-11 15:55
82% SWE Bench is really not a small figure, and the key is that it hasn't hit the ceiling yet. This momentum is quite intense.
View OriginalReply0
DeFiChef
· 12-11 15:55
82%? That's a pretty impressive number. Feels like the ceiling hasn't even been reached yet.
View OriginalReply0
SigmaBrain
· 12-11 15:32
82% is really impressive, but I'm more concerned about whether this trajectory can be maintained, and not get stuck at 90%.
Four months have passed since the initial rollout, and the 5.1 iteration brought subtle refinements rather than a complete overhaul. The team skipped the playful codename this time—unlike previous naming conventions—because this feels different. We're witnessing a genuine leap forward here.
Hitting 82% on SWE Bench isn't just impressive; the trajectory suggests there's still room to climb. No plateau in sight yet.