What is Cursor?
Cursor is an AI-powered code editor that includes the Composer 2.5 coding model. Its trajectory has been marked by rapid growth, acquisition, and competitive positioning against frontier models.
Release history
- Jan 2026 – CEO Michael Truel told staff it was “wartime,” recognizing Cursor’s business model was being eroded from both sides, per Nathaniel Whittemore.
- Apr 2026 – Harry Stebbings reported that Cursor was the biggest private acquisition ever in venture, with a roughly break-even gross margin business.
- May 2026 – Cursor scored 69.3% on Terminal Bench 2.0, behind Opus 4.7, and performed comparably to Opus 4.7 and GPT-5.5 on SweBench multilingual at 10–60x lower cost, per Nathaniel Whittemore.
- Jun 2026 – Martin Casado noted Cursor’s run rate had grown from $9 billion at end of last year to $47 billion; Jason Lemkin stated Cursor was being purchased for $60 billion; Nathaniel Whittemore reported a new Cursor model teased at Compile, same size as Claude Opus and GPT-5.5, trained from scratch with no more Kimi base, using 10–20x more compute versus Composer, generally intelligent not just coding, releasing in the next couple of weeks; Ronak Malde noted that binary feedback signals like accept/reject in Cursor are noisy, as users rarely give explicit feedback and often “vibe accept” everything.
- Aug 2026 – Harry Stebbings said Cursor is “going to become a GitHub entire workflow replacement in a couple months,” and that Elon “swooped in” to buy Cursor, which pushed it up his priority list. Simon Mo noted a healthy ecosystem development, citing MiniMax’s M2.7 release with a term focusing on usage and derivative works, and mentioned Fireworks and Cursor in that context.
In the discourse
Attributed discussion of Cursor.
Cursor's Composer 2.5 scores 69.3% on Terminal Bench 2.0 and is comparable to Opus 4.7 (80.5%) and GPT-5.5 (77.8%) on SweBench multilingual.
“It scored 69.3% on Terminal Bench 2.0, which is just behind Opus 4.7 On SweBench multilingual, it scored comparable to both Opus 4.7 at 80.5% and GPT-5.5 at 77.8%.”Nathaniel Whittemore · 20 May 2026
A break-even gross margin business achieved a $60B exit, directly contradicting the SaaS doctrine that high gross margins are prerequisite for large software exits.
“This is a break even gross margin business but it worked because the overarching comment is AI is the biggest story of the last 5 years and big tech wants to be a player.”Harry Stebbings · 23 Apr 2026
Cursor's $60B exit is the largest private acquisition in venture history, surpassing Wiz and WhatsApp.
“This is the biggest private acquisition ever in venture.”Harry Stebbings · 23 Apr 2026
Cursor's annualized run rate grew from $9 billion to $47 billion in a short period, per Martin Casado.
“When swap it's gone from whatever it was 9 billion run rate at the end of last year to $47 billion run rate.”Martin Casado · 8 Jun 2026
Cursor CEO Michael Truel declared internal 'wartime' in January as the company's business model faced erosion from both sides.
“In January, CEO Michael Truel told staff that it was, quote, wartime, recognizing that Cursor's business model was being eroded from both sides.”Nathaniel Whittemore · 20 May 2026
Harry Stebbings on the tiny pool of acquirers able to write a $60B check for a break-even business.
“The number of people who can write a $60 billion check for a roughly break even gross margin business can be counted on the fingers of one hand removing at least one or two of them.”Harry Stebbings · 23 Apr 2026
Cursor Composer 2.5 matched or approached Opus 4.7 and GPT 5.5 performance at 10 to 60x lower cost, making it a significant efficiency play for enterprise coding workloads.
“It performed at a very comparable level to Opus 4.7 and GPT 5.5, a little ahead of them on their medium settings and a little below them on extra high and max settings, but it did all that at 10 to 60x lower cost.”Nathaniel Whittemore · 24 May 2026
Cursor Composer 2.5 achieved comparable performance to Opus 4.7 and GPT 5.5 at 10 to 60x lower cost.
“It performed at a very comparable level to Opus 4.7 and GPT 5.5, a little ahead of them on their medium settings and a little below them on extra high and max settings, but it did all that at 10 to 60x lower cost.”Nathaniel Whittemore · 24 May 2026
MiniMax M2.7 and Kimi K3 signal a shift in open-weight licensing toward usage-based restrictions, moving away from permissive Apache-style terms.
“Now we're seeing a very healthy ecosystem development starting from even miniax recently when they're releasing the M2.7 model they have a term specifically focusing on usage and ki initially also has like if you have derivative works like this is kind of big news back then was fireworks and cursor about how they built on top of Kimi model.”Simon Mo · 6 Aug 2026
Cursor targets a shortlist of roughly 50 best-in-world candidates per role rather than running a high-volume funnel.
“Here's the 50 people that would be the best 50 in the world for this. We just need one of these 50.”Adam Ward · 9 Aug 2026