eric zakariasson(@ericzakariasson)

try grok 4.5 in cursor now! double usage for the first week included. its a good model

eric zakariasson (@ericzakariasson) — Wed, 08 Jul 2026 18:01:00 GMT

try grok 4.5 in cursor now! double usage for the first week included. its a good model

Cursor@cursor_ai

We've partnered with SpaceXAI to train Grok 4.5.

It’s our most powerful model yet and the first we've built for more than software engineering.

💬152🔄119❤️2623👀263339📊421

https://t.co/3rx8RZj6Mr

eric zakariasson (@ericzakariasson) — Fri, 03 Jul 2026 11:49:30 GMT

x.com/i/article/2072…

💬18🔄21❤️291👀36166📊109

https://t.co/9m2Ia1cvJg

eric zakariasson (@ericzakariasson) — Thu, 02 Jul 2026 12:50:25 GMT

x.com/ericzakariasso…

eric zakariasson@ericzakariasson

x.com/i/article/2072…

💬5🔄0❤️13👀15499📊6

https://t.co/WasA5WdOao

eric zakariasson (@ericzakariasson) — Thu, 02 Jul 2026 11:10:50 GMT

x.com/i/article/2072…

💬39🔄56❤️1086👀121632📊391

fable is so back, use it again in cursor!

eric zakariasson (@ericzakariasson) — Wed, 01 Jul 2026 20:20:14 GMT

fable is so back, use it again in cursor!

Cursor@cursor_ai

Claude Fable 5 is available again in Cursor.

It leads all models on CursorBench, but is the most expensive per task.

💬27🔄6❤️499👀38341📊65

https://t.co/GDtbRoZZnC

eric zakariasson (@ericzakariasson) — Mon, 29 Jun 2026 17:07:22 GMT

apps.apple.com/app/cursor/id6…

💬2🔄0❤️25👀11410📊5

i've been using cursor mobile on the go for the last weeks, and having access to all cloud agents fr...

eric zakariasson (@ericzakariasson) — Mon, 29 Jun 2026 17:06:35 GMT

i've been using cursor mobile on the go for the last weeks, and having access to all cloud agents from everywhere is really nice

go on a walk, get an idea, dictate it in the app
come back from walk to a finished agent where you can jump into it

try it today!

Cursor@cursor_ai

Introducing Cursor for iOS.

Build from anywhere by launching always-on cloud agents. Or remotely control agents running on your computer from the app.

Composer 2.5 is 75% off in the app now through July 5.

Your browser does not support the video tag.

💬200🔄609❤️2250👀962681📊536

its going to be a good model

eric zakariasson (@ericzakariasson) — Mon, 29 Jun 2026 08:07:08 GMT

its going to be a good model

Elon Musk@elonmusk

In addition to their excellent and unique training data, the Cursor team is also making major engineering contributions to v9 SFT & RL. It’s an honor and a pleasure to work with them. 

For this 1.5T run, Cursor data was added in supplemental training, which is not quite as good as having it in initial training.

The 2T run that started a few weeks ago has greatly improved data in scope & scale in almost every arena, and many upgrades/fixes to the training recipe. That will finish in late July for August release.

💬25🔄13❤️835👀62799📊94

https://t.co/drBAUlV6B3

eric zakariasson (@ericzakariasson) — Fri, 26 Jun 2026 13:04:23 GMT

x.com/i/article/2070…

💬29🔄77❤️1132👀115433📊533

model agnostic agent in slack

eric zakariasson (@ericzakariasson) — Wed, 24 Jun 2026 07:28:34 GMT

model agnostic agent in slack

Gergely Orosz@GergelyOrosz

Interesting to observe Anthropic going from the moat being the best model to building a tooling ecosystem with right integrations to common dev + non-dev workflows.

If I was a CTO I’d only have a Slack integration where I can switch models *anytime*… to avoid lock-in tho

💬21🔄15❤️741👀81880📊94

stay tuned for next model

eric zakariasson (@ericzakariasson) — Tue, 23 Jun 2026 07:43:45 GMT

stay tuned for next model

Cursor@cursor_ai

Three announcements from our keynote at Compile, including how we're training a new model with SpaceX.

Your browser does not support the video tag.

💬85🔄37❤️1477👀89807📊210

mobilemaxxing cursor (app is soon GA)

eric zakariasson (@ericzakariasson) — Wed, 17 Jun 2026 22:40:18 GMT

mobilemaxxing cursor (app is soon GA)

Cursor@cursor_ai

It’s now easier to move local agents to the cloud so they can keep working with your laptop closed.

Prompt Cursor from your phone, run many agents in parallel, and get back PRs with demos of their work.

Your browser does not support the video tag.

💬60🔄11❤️512👀63441📊102

extremely excited to join this talented team. a lot of things in the works!

eric zakariasson (@ericzakariasson) — Tue, 16 Jun 2026 16:28:51 GMT

extremely excited to join this talented team. a lot of things in the works!

Cursor@cursor_ai

We're excited to join forces with @SpaceX to advance the frontier of useful AI. Expect significant improvements to Cursor soon.

💬54🔄8❤️963👀57138📊126

what do you have in your dock?

eric zakariasson (@ericzakariasson) — Mon, 15 Jun 2026 21:44:01 GMT

what do you have in your dock?

💬164🔄9❤️510👀87568📊234

mcp > cli im glad this debase is not as active anymore. i use both every day. they just serve diffe...

eric zakariasson (@ericzakariasson) — Mon, 15 Jun 2026 17:15:52 GMT

mcp > cli

im glad this debase is not as active anymore. i use both every day. they just serve different purposes

cli for stuff the model already knows. git, gh, npm, docker, file ops. trained on man pages, and costs almost nothing in context. if im already signed in locally theres no reason to wrap it in anything

mcp for most integrations. slack, notion, linear, twitter

and its neat to have a protocol for all these integrations
- add one server to my teams cursor and everyone gets access
- auth once, persists, same locally and in cloud

it also just feels better in cursor. rich icons, traceable, you can easily follow whats happening (tbh not reading that much)

cli for personal, mcp for team (with oauth)

💬63🔄29❤️532👀69079📊179

see you at compile!

eric zakariasson (@ericzakariasson) — Sat, 13 Jun 2026 21:49:01 GMT

see you at compile!

Lee Robinson@leerob

See you at Cursor Compile next Tuesday in SF!

I'll be talking with @levelsio about retro computing, building your ideas, lifting, the perfect steak, and more.

💬10🔄0❤️104👀27188📊17

cafe cursor sf just opened and its already a long long line (2000 signups) excited to see you all!

eric zakariasson (@ericzakariasson) — Wed, 10 Jun 2026 17:21:04 GMT

cafe cursor sf just opened and its already a long long line (2000 signups)

excited to see you all!

💬29🔄3❤️380👀29345📊74

we just shipped some improvements to https://t.co/Mmm7vHWudD! you can now see cost, output tokens a...

eric zakariasson (@ericzakariasson) — Tue, 09 Jun 2026 17:49:22 GMT

we just shipped some improvements to cursor.com/evals!

you can now see cost, output tokens and steps plotted in the graph for each model

Your browser does not support the video tag.

nate@_nateschmidt

cursor.com/evals now includes steps and output tokens as well! These are additional signals our team uses to eval models.

💬22🔄21❤️441👀63647📊72

go try out fable in cursor, it's an incredible but expensive model!

eric zakariasson (@ericzakariasson) — Tue, 09 Jun 2026 17:12:17 GMT

go try out fable in cursor, it's an incredible but expensive model!

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.

💬39🔄10❤️521👀56731📊81

3. Dependency-upgrade grinder /loop through every outdated package one at a time: bump it, run buil...

eric zakariasson (@ericzakariasson) — Mon, 08 Jun 2026 23:08:18 GMT

3. Dependency-upgrade grinder

/loop through every outdated package one at a time: bump it, run build, type-check, and tests, fix whatever breaks, commit, then move to the next. Keep going until the list is empty and everything's green.

💬5🔄2❤️62👀9517📊21

2. Post-deploy error watch I just shipped [thing]. /loop every 5m polling my error tracker and logs...

eric zakariasson (@ericzakariasson) — Mon, 08 Jun 2026 23:08:15 GMT

2. Post-deploy error watch

I just shipped [thing]. /loop every 5m polling my error tracker and logs for new error groups on this service, and for each new one triage it and either push a hotfix or flag for rollback. Fall back to a heartbeat but wake immediately if a new error spikes.

💬1🔄0❤️49👀11382📊12

here are 3 loops you can run in cursor 1. Flaky-test exterminator /loop run my test suite 20 times...

eric zakariasson (@ericzakariasson) — Mon, 08 Jun 2026 23:08:12 GMT

here are 3 loops you can run in cursor

1. Flaky-test exterminator

/loop run my test suite 20 times, collect every intermittent failure, fix or quarantine the flaky ones, and don't stop until you get 5 consecutive fully-green runs.

💬20🔄24❤️558👀58616📊184

>npx --yes was-gpt-here@latest >yes

eric zakariasson (@ericzakariasson) — Sat, 06 Jun 2026 00:53:34 GMT

>npx --yes was-gpt-here@latest
>yes

Your browser does not support the video tag.

Christoph Nakazawa@cnakazawa

You are not tokenmaxxing if you don't have at least 20 copies of this function in your repo.

💬3🔄1❤️81👀20104📊14

design mode must know shortcuts: Toggle: ⌘⇧D Tab → next sibling Shift+Tab → prev sibling Enter → f...

eric zakariasson (@ericzakariasson) — Fri, 05 Jun 2026 18:13:43 GMT

design mode must know shortcuts:

Toggle: ⌘⇧D

Tab → next sibling
Shift+Tab → prev sibling
Enter → first child
Shift+Enter → parent

Add to chat: ⌘L

Drag → box/scribble
Shift+click → multi-select
Alt+click → add to chat

Your browser does not support the video tag.

Cursor@cursor_ai

With Design Mode, you can now point, draw, or talk to update your UI.

Your browser does not support the video tag.

💬21🔄7❤️286👀29814📊69

if you're in a team, you can enable public profiles from here: https://t.co/3wstPh7C7K

eric zakariasson (@ericzakariasson) — Thu, 04 Jun 2026 17:28:26 GMT

if you're in a team, you can enable public profiles from here: cursor.com/dashboard/team…

💬1🔄0❤️32👀13206📊6

we're building a lot more into profiles, and will make your handle useful across cursor surfaces ov...

eric zakariasson (@ericzakariasson) — Thu, 04 Jun 2026 17:17:53 GMT

we're building a lot more into profiles, and will make your handle  useful across cursor surfaces over time.

here's my profile: cursor.com/@eric

💬19🔄2❤️120👀41214📊37

introducing cursor profiles! go claim your handle at https://t.co/6t5lg2jqvg

eric zakariasson (@ericzakariasson) — Thu, 04 Jun 2026 17:17:50 GMT

introducing cursor profiles!

go claim your handle at cursor.com/profile

Your browser does not support the video tag.

💬355🔄97❤️1823👀715496📊813

https://t.co/0PIwVLwUrC

eric zakariasson (@ericzakariasson) — Wed, 03 Jun 2026 15:45:36 GMT

x.com/i/article/2061…

💬25🔄42❤️738👀114018📊323

cursor in slack can now read documents attached in the thread, including .txt, .log, .json, .zip, .p...

eric zakariasson (@ericzakariasson) — Tue, 02 Jun 2026 23:41:06 GMT

cursor in slack can now read documents attached in the thread, including .txt, .log, .json, .zip, .pdf, or .docx files!

Your browser does not support the video tag.

💬13🔄17❤️266👀19047📊55

lmao all of our designers got one of their apple displays replaced by the cheapest one from amazon ...

eric zakariasson (@ericzakariasson) — Tue, 02 Jun 2026 05:01:31 GMT

lmao all of our designers got one of their apple displays replaced by the cheapest one from amazon

in pursuit of higher contrast

💬84🔄30❤️2287👀367457📊403

i played with this way too long. thanks for building it @DanHollick https://t.co/2l1Fu2Qg2M

eric zakariasson (@ericzakariasson) — Mon, 01 Jun 2026 22:27:06 GMT

i played with this way too long. thanks for building it @DanHollickcursor.com/insights

Your browser does not support the video tag.

💬4🔄5❤️188👀10457📊34

wow i didnt know they made cyberlimos like this

eric zakariasson (@ericzakariasson) — Sat, 30 May 2026 16:56:02 GMT

wow i didnt know they made cyberlimos like this

💬80🔄56❤️2169👀191012📊354

based on this gem from @AlexWauters https://t.co/TF4N8GcHcg

eric zakariasson (@ericzakariasson) — Fri, 29 May 2026 17:31:23 GMT

based on this gem from @AlexWautersscalex.dev/blog/ai-agent-…

💬0🔄0❤️11👀3512📊3

play this game to see how many terminal commands you can safely approve in 30 seconds. try to beat a...

eric zakariasson (@ericzakariasson) — Fri, 29 May 2026 17:31:20 GMT

play this game to see how many terminal commands you can safely approve in 30 seconds. try to beat auto review, it's harder than you think!

cursor-auto-review.com

Your browser does not support the video tag.

Cursor@cursor_ai

Auto-review mode is now available in Cursor.

It allows agents to run tool calls with fewer approval prompts and safer execution.

Your browser does not support the video tag.

💬16🔄9❤️314👀43968📊55

grok build is available in cursor, try it out!

eric zakariasson (@ericzakariasson) — Fri, 29 May 2026 12:57:01 GMT

grok build is available in cursor, try it out!

💬86🔄35❤️1128👀68905📊229

it's composable so you can run a single skill (just security or just code quality) or both in parall...

eric zakariasson (@ericzakariasson) — Thu, 28 May 2026 17:58:16 GMT

it's composable so you can run a single skill (just security or just code quality) or both in parallel via /thermos

same building blocks, different scopes

install with `/add-plugin thermos` or from marketplace

💬0🔄2❤️56👀3875📊14

audits are diff-first. only the diff, no rabbit-holes into pre-existing code the security audit onl...

eric zakariasson (@ericzakariasson) — Thu, 28 May 2026 17:58:12 GMT

audits are diff-first. only the diff, no rabbit-holes into pre-existing code

the security audit only reads PR or bugbot threads after its independent audit lands. fresh eyes, no anchoring

when both reviewers flag the same issue, it ranks higher. overlap is signal!

💬1🔄0❤️18👀4189📊3

introducing thermos in cursor a deep security/correctness audit and a harsh code quality audit, run...

eric zakariasson (@ericzakariasson) — Thu, 28 May 2026 17:58:02 GMT

introducing thermos in cursor

a deep security/correctness audit and a harsh code quality audit, run in parallel on your branch, synthesized into one prioritized list

Your browser does not support the video tag.

💬34🔄61❤️1049👀51713📊300

s/o @DevangJhabakh for building this!

eric zakariasson (@ericzakariasson) — Thu, 28 May 2026 16:03:50 GMT

s/o @DevangJhabakh for building this!

💬2🔄0❤️7👀2661📊3

the economics of intelligence > Cost per agent request varies by nearly 9x across model families, s...

eric zakariasson (@ericzakariasson) — Thu, 28 May 2026 16:01:28 GMT

the economics of intelligence

> Cost per agent request varies by nearly 9x across model families, showing that the same workflow can have very different cost profiles depending on the model behind it.

some higher cost models are cheaper in the long run due to increased intelligence, but for p50 requests a model like composer 2.5 will do the job both faster and cheaper

a lot of interesting data in this report, recommend a read

Cursor@cursor_ai

Introducing the Cursor Developer Habits Report.

We’re sharing some of our findings on how software development is changing.

It’s based on the most comprehensive dataset on AI coding in the world, across all model families.

Your browser does not support the video tag.

💬15🔄16❤️262👀27316📊47

you can now share canvases in cursor! also, you can use design mode to make sniper edits in your can...

eric zakariasson (@ericzakariasson) — Thu, 28 May 2026 05:39:52 GMT

you can now share canvases in cursor! also, you can use design mode to make sniper edits in your canvases. very in flow with the speed of composer 2.5

here's a paperclipmaxxing canvas i just built and shared using slack mcp

Your browser does not support the video tag.

💬30🔄12❤️380👀37042📊100

its so fun seeing all projects built with cursor sdk that leverages the composer 2.5 speed. more gen...

eric zakariasson (@ericzakariasson) — Wed, 27 May 2026 04:47:02 GMT

its so fun seeing all projects built with cursor sdk that leverages the composer 2.5 speed. more generative, synchronous experiences for users when it comes with really high tps

💬26🔄1❤️160👀11484📊44

what do you think of composer 2.5 so far? how can we make the next model even better? want to hear ...

eric zakariasson (@ericzakariasson) — Mon, 25 May 2026 17:37:55 GMT

what do you think of composer 2.5 so far? how can we make the next model even better?

want to hear your feedback on behavior, speed, quality, whatever!

💬435🔄20❤️1161👀111740📊580

hey @grok come up with 10 examples of what i can build with the cursor sdk over the weekend (its 90%...

eric zakariasson (@ericzakariasson) — Fri, 22 May 2026 23:48:38 GMT

hey @grok come up with 10 examples of what i can build with the cursor sdk over the weekend (its 90% off). here's a cookbook if that helps github.com/cursor/cookboo…

💬23🔄5❤️255👀48338📊95

90% off composer 2.5 this weekend in cursor sdk go build something cool and post it here!

eric zakariasson (@ericzakariasson) — Fri, 22 May 2026 19:59:01 GMT

90% off composer 2.5 this weekend in cursor sdk

go build something cool and post it here!

Cursor@cursor_ai

With the Cursor SDK, you can build your own agents with Composer 2.5. It's now available in Python and TypeScript.

This long weekend, Composer usage is 90% off in the SDK. We're excited to see what you build!

Your browser does not support the video tag.

💬42🔄18❤️480👀41106📊93

try it here https://t.co/8R4XNCUOfe

eric zakariasson (@ericzakariasson) — Thu, 21 May 2026 17:58:17 GMT

try it here cursor.com/marketplace/cu…

💬4🔄17❤️488👀40762📊247

the most used skill internally at cursor right now /thermo-nuclear-code-quality-review - deletes c...

eric zakariasson (@ericzakariasson) — Thu, 21 May 2026 17:58:14 GMT

the most used skill internally at cursor right now

/thermo-nuclear-code-quality-review

- deletes complexity instead of moving it
- blocks files over 1k lines
- flags thin wrappers and leaked logic
- rejects PRs that work but make code messier

Your browser does not support the video tag.

💬136🔄260❤️5444👀469544📊2178

i wrote a guide on optimizing context usage 6 months ago that i never posted. back then with the mod...

eric zakariasson (@ericzakariasson) — Thu, 21 May 2026 13:02:04 GMT

i wrote a guide on optimizing context usage 6 months ago that i never posted. back then with the models available, you could only pick 2 of 3:

1. intelligent
2. fast
3. cheap

intelligent + fast = expensive
fast + cheap = dumb
cheap + intelligent = slow

now, with composer 2.5, this is no longer true and the post is obsolete. looking at TPS, avg cost per task, and score from cursorbench, it's clearly capable of all three

but benchmarks are just benchmarks. what matters is how it feels to use and if it can actually accomplish your tasks. from the feedback so far, that's very much the case

go try it out if you haven't already

Artificial Analysis@ArtificialAnlys

Cursor's new Composer 2.5 takes third on the Artificial Analysis Coding Agent Index and is ~10-60x lower cost than the higher-effort Opus 4.7 and GPT-5.5 variants above it. This release puts Composer among the leading coding agent models, something that wasn’t clear for past releases

@cursor_ai has released Composer 2.5, the latest model in its Composer line. Composer 2.5 scored 62 on our Coding Agent Index, a 14 point gain over Composer 2 (48). This puts it in third place of our tested agents, behind only Claude Opus 4.7 (max) in Claude Code (66) and GPT-5.5 (xhigh reasoning) in Codex (65). These cost $4.10 and $4.82 per task respectively, ~10x the cost of Composer 2.5 Fast ($0.44) and ~60x the cost of Composer 2.5 standard ($0.07).

Key results for Composer 2.5 in Cursor CLI:

➤ Cost-quality Pareto frontier: At $0.07 (standard) and $0.44 (Fast) per task, Composer 2.5 is cheaper than every other agent scoring above 60 on the Index. Medium-effort peers cost $1.24–$2.21 per task; higher-effort variants land 3-4 points above at $4.10–$4.82

➤ Per-benchmark gains vs Composer 2: +35 points on SWE-Bench-Pro-Hard-AA (12% → 47%), +2 points on Terminal-Bench v2 (64% → 66%), and +3 points on SWE-Atlas-QnA (69% → 72%). At 47%, Composer 2.5's score on SWE-Bench-Pro-Hard-AA is comparable to Claude Opus 4.7 (max) in Claude Code

➤ Among the fastest coding agents: Composer 2.5 Fast runs at an average wall time of 6.7 minutes per task, the third-fastest agent on the Artificial Analysis Coding Agent Index, behind only Claude Opus 4.7 (medium) in Claude Code (5.8m) and GPT-5.5 (medium) in Cursor CLI (6.2m)

➤ Fast mode enables better responsiveness at 6x pricing: Fast runs 30% faster than standard Composer 2.5, but is ~6x the cost per task ($0.44 vs $0.07). Token pricing is 6x higher for Fast: $3.00/$15.00 vs $0.50/$2.50 per million input/output tokens

Model details:

➤ Base model: Continued training on @Kimi_Moonshot's open weights Kimi K2.5 as with Composer 2, with Cursor reporting ~85% of total compute from its own additional training and reinforcement learning

➤ Pricing: $0.50/$2.50 per million input/output tokens for the standard variant; $3.00/$15.00 for the Fast variant (the default in Cursor)

➤ Available exclusively in Cursor: both Cursor IDE and Cursor CLI, an externally accessible API is not available

Congratulations @cursor_ai and @mntruell on the impressive release!

💬29🔄12❤️349👀39094📊72