Original post
Super Dario#1958
Super Dario#1958Cognition integrates Claude Fable 5 into Devin, topping its new FrontierCode benchmark with a 46.3 percent score
Story Overview
Anthropic's Claude Fable 5 variant has landed inside Cognition's Devin agent and taken the lead on the company's new FrontierCode benchmark, which grades models on whether their code actually merges cleanly into real projects instead of just passing isolated tests.
1202K120195188.4K
Developer Impact
Coding agents get a quality nudge
The integration lets Devin users tap Fable 5 for engineering tasks right now, though rollout details and exact performance numbers remain tied to Cognition's announcement.
Open Question
New eval raises fresh comparison questions
FrontierCode's emphasis on mergeability and overall quality is still early, so it is unclear how widely teams will adopt it versus existing coding benchmarks.
Sentiment
Many users celebrated Claude Fable 5's large benchmark lead after Devin integration for its doubled scores and widening gaps over rivals, while others dismissed the results as self-serving since Cognition designed the test.
Pos
60.3%
Neg
39.7%
25 comments with sentiment.
Cluster Engagement
Posts from X
Most Activity
Most Activity
VIEWS31.8KBOOKMARKS28LIKES352
Nick Dobos@NickADobos
Took 1 day for AI to 2x the score on the hardest programming benchmark ever made
Scott Wu@ScottWu46
@NickADobos I hope that it gets solved in the next 6 months and then we can move on to even more challenging tasks!
1dViews 31.8KLikes 352Bookmarks 28
RETWEETS10REPLIES11
Scott Wu@ScottWu46
A new top scorer just one day after our benchmark released! Especially strong on the hardest tasks: 13.4% -> 29.3% on FrontierCode Diamond compared to Opus 4.8.
Cognition@cognition
Claude Fable 5 is now available in Devin.
Fable 5 earns the #1 spot on FrontierCode, our benchmark for real-world engineering tasks that grades mergeability and quality:
1dViews 14.8KLikes 212Bookmarks 11
Cognition@cognition
You can try Claude Fable 5 as part of Devin Cloud’s Ultra agent. Devin Ultra is our smartest and most capable agent, which excels at long-horizon tasks and debugging.
We tuned the harness so Ultra costs only ~40% more than default Devin agent.
Claude Fable 5 is also available in Devin Desktop and Devin CLI.
1dViews 3.3KLikes 40Bookmarks 3
Cognition@cognition
Try out Devin today! http://devin.ai
1dViews 1.8KLikes 10
Joe Lonsdale@JTLonsdale
@cognition 🔥
1dViews 2.1KLikes 9
Tim Jayas@TimJayas
@cognition OpenAI needs to cook something big now to compete 💀
1dViews 700Likes 4
Cognition@cognition
Read our blog post: https://devin.ai/blog/claude-fable-5-available-in-devin
1dViews 953Likes 5
Jeremy@WonsonYo
@cognition @noahzweben I’m scared to try it, run out of my Devin max creds immediately as is 👀
1dViews 297Likes 1
Alex Kaplan@alexkaplan0
@cognition Are we already gonna get saturated 😭
1dViews 288Likes 5
Evgeniy Trifonov@phillennium
@NickADobos The benchmark creators obviously had pre-release Fable access, so it seems it was purposefully published a day ago for such an effect
22hViews 172Likes 1
luke walksky@wussupitsluke
@cognition Gaslighting benchmarks. Real world 5.5 mogs 4.8.
1dViews 414Likes 2
Terp@OnlyTerp
@cognition LETS GOO
1dViews 234Likes 2
Jakub Hojsan@JakubHojsan
@ScottWu46 wonder where next OAI model will land
17hViews 17
Quantum Voyager@AgiWhen
@cognition when are you guys testing the minimax M3?
1dViews 485Likes 1
Behnam@OrganicGPT
@cognition @noahzweben bro you can't be serious about Fable scoring almost x2 of gpt 5.5
1dViews 378Likes 1
andrew gao@itsandrewgao
@cognition https://cognition.ai/blog/frontier-code
1dViews 319Likes 1
Chase Dalton@ChaseHDalton
@cognition Mirror, Mirror On the Wall, Who is the Mightiest Model of Them All...
1dViews 864
Carbon@Carbon19992025
@cognition @devarbol definitely not a bubble. Another step change in capability.
1dViews 7
Ruhul Amin@ruhul_swe
@cognition Finally, Devin's user will get a taste of Fable 5
1dViews 187Likes 1
Omri Casspi עומרי משה כספי@Casspi18
@cognition 💪🏽
23hViews 517
Load more posts