Cursor Acknowledges Its New Low-Cost Coding Model Has Chinese Roots

Cursor simply acknowledged that its latest coding model has Chinese language roots — a element it not noted the primary time round.

In a sequence of posts on X over the weekend, Cursor executives mentioned Composer 2 was initially constructed on high of Kimi K2.5, an open-source mannequin developed by Chinese startup Moonshot AI.

“We have evaluated quite a lot of base fashions on perplexity-based evals and Kimi k2.5 proved to be the strongest!” mentioned Cursor’s cofounder Aman Sanger on X on Saturday.

“It was a miss to not point out the Kimi base in our weblog from the beginning,” he added.

The disclosure seems to have been sparked by an X consumer named Fynn, who posted on Friday that Composer 2 was “simply Kimi 2.5” with extra reinforcement studying.

To assist the declare, the consumer pointed to code snippets that appeared to reference Kimi because the underlying system.

‘Not less than rename the mannequin ID,” the consumer wrote.

In response to the consumer’s X put up, Cursor’s vp of developer schooling, Lee Robinson, acknowledged that Composer 2 was constructed on Kimi K2.5 as an open-source base.

“We are going to do full pretraining sooner or later,” Robinson mentioned.

“Solely ~1/4 of the compute spent on the ultimate mannequin got here from the bottom, the remaining is from our coaching,” he added.

Robinson additionally mentioned the corporate is complying with the mannequin’s licensing phrases via its inference supplier.

The Chinese language startup posted on X on Saturday that Cursor is utilizing Kimi K2.5 underneath a certified industrial partnership.

“Seeing our mannequin built-in successfully via Cursor’s continued pretraining & high-compute RL coaching is the open mannequin ecosystem we like to assist,” the put up learn.

Cursor was final valued at $29.3 billion in November.

Cursor’s new mannequin is cheaper and higher

Cursor mentioned in a weblog put up on Thursday that Composer 2 is “frontier-level at coding” and priced at $0.50 per million enter tokens and $2.50 per million output tokens, calling it “a brand new, optimum mixture of intelligence and value.”

By comparability, Anthropic’s Claude Opus 4.6 is priced at $5 per million enter tokens and $25 per million output tokens, whereas Claude Sonnet 4.6 prices $3 and $15, respectively, in response to the corporate’s web site.

That places Composer 2 at roughly one-tenth the price of Opus 4.6 and about one-sixth the price of Sonnet 4.6 on each enter and output tokens.

Customers on X have added to the controversy, with some praising the efficiency of Kimi after studying that Composer 2 was built on top of it.

“As somebody who mainly lives in opus 4.6, seeing an open-weight kimi 2.5 fine-tune truly beat it on coding benchmarks is wild,” one X consumer wrote in response to Fynn’s put up.

“Properly that is an indication for RL Chinese language is in new recreation,” one other consumer wrote, referring to reinforcement studying.

Others had been extra vital of Cursor’s dealing with of the disclosure, questioning why the corporate didn’t acknowledge Kimi upfront.

“Cursor is changing into a mannequin routing layer, not an IDE. they choose the most affordable mannequin that clears a high quality bar per activity, wrap it of their UX, and pocket the margin,” one consumer who goes by aira wrote on X.