Original post
Marco Mascorro#558
Marco Mascorro#558
Cursor@cursor_ai
Claude Fable 5 is now available in Cursor.
It sets a new state of the art on CursorBench at 72.9%, 8 points above the previous best.
10:10 AM · Jun 9, 2026 · 302.2K Views
Sentiment
Positive users praise Claude Fable 5's record CursorBench score and new evals features for better productivity, while negative users object to Cursor's high subscription costs and demand a mobile app instead.
Pos
54.4%
Neg
45.6%
47 comments with sentiment.
Cluster Engagement
Posts from X
Most Activity
Most Activity
VIEWS16.6KBOOKMARKS6LIKES78RETWEETS5
Cursor@cursor_ai
See how Claude Fable 5 compares across every model: http://cursor.com/evals
3hViews 16.6KLikes 78Bookmarks 6
REPLIES7
Jacob Beers@JTBeers
@cursor_ai Why is the x axis backwards?
3hViews 4.4KLikes 62
Every 📧@every
@cursor_ai We've been testing it for a week. Here are the best workflows: https://every.to/vibe-check/anthropic-mythos-our-fable-vibe-check
3hViews 723Likes 9Bookmarks 3
piet@iampiet
@cursor_ai the more things change the more they stay the same
3hViews 2.1KLikes 26
Ali Romman@aliromman_
@cursor_ai The benchmarks are insane.
3hViews 2.2KLikes 6Bookmarks 1
Sami Hindi@DevBySami
@cursor_ai Use 5.5 high. Not fair to compare high to medium
3hViews 3.1KLikes 12
lost in latency@lostinlatencyX
@cursor_ai composer 2.5 is close to fable 5's lowest is what impresses me 😅
3hViews 1.8KLikes 4Bookmarks 1
Leon Lin@LexnLin
@ericzakariasson is low effort good enough? 😭
3hViews 485Likes 5
Nischal@NischalSkanda
@cursor_ai This tells me everything that I needed to know! you guys are the best.
2hViews 504Likes 4Bookmarks 1
Sundaram Kumar Jha@skjtwts
@cursor_ai we need composer 3, Fable 5 cost api cost is too high
3hViews 1KLikes 7
Andrew | Clippster.app@andrew_da_miz
@cursor_ai I really like how gpt 5.5 xhigh is conveniently missing
2hViews 776Likes 3
Denis Spirin@den_spirin
@cursor_ai That is great, but I have to ask, shouldn't the default be "medium," judging by the graph?
That seems like a more reasonable default, given the bang for the buck
2hViews 1.5KLikes 1
Ryan Marcus@RyanMarcus43627
@cursor_ai Is the ZDR model consent popover missing from the dashboard? I cannot find a way to enable it for my team.
2hViews 866Likes 1
Pipc@Pipc2025
@cursor_ai Get $400 of API credit for Claude FABLE 5 for only $100 :) https://cursor.com/referral?code=ZLYGHVFTMSOE
3hViews 229Likes 2Bookmarks 1
Dave@Uncle__Dave
@cursor_ai 72.9% on CursorBench is nice but the 8 point jump is the part I'd actually want to see hold up on real repos, not just the benchmark. half my Claude wins evaporate the moment there's a weird monorepo path involved.
2hViews 609Likes 1Bookmarks 1
Dev K@DesiRichDev
@cursor_ai everyone's announcing Fable 5 integration today. none of them built the model. the real question is who owns the workflow when the model doesn't matter anymore.
2hViews 1.5KBookmarks 1
B@RĮ$@Baris7722
@cursor_ai 72.9% on CursorBench—a full 8-point leap. This isn't just a benchmark win; it's a signal that AI coding assistants are moving from "helpful" to indispensable for dev workflows.
3hViews 92Likes 2
nsxdavid@nsxdavid
@cursor_ai Are you going to pull it from Cursor plans on the 23rd as well?
3hViews 3.1KLikes 2
gary IH fung@garyfung
@cursor_ai where's Composer Pro? Please draw a line extending upward and to the left from the 2.5 dot
3hViews 708Likes 3
niko@0x_nik0
@cursor_ai 8 point jump on cursorbench is wild - wonder if that holds on real codebases
3hViews 1.5KLikes 2
Load more posts