Unverified50% confidenceOpinionExact time
Cursor's next-generation in-house model Composer 3, internally codenamed Vega, is rumored in the community to have been fully pre-trained from scratch with a scale exceeding 1.5 trillion parameters.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
9/12/2026
First Seen
Valid until: 12/11/2026
Sources
Related Entities
Related Claims
UnverifiedCursor's Composer 2 is trained based on Kimi 2.5, which is a Mixture-of-Experts model with 1 trillion parameters and 30 billion active parameters.72% similarUnverifiedCursor's Composer series of models uses real-time online RL updated every 5 hours, based on continuous pre-training on hundreds of trillions of tokens.69% similarUnverifiedCursor 2.0 released its self-developed Composer model, no longer relying entirely on third-party foundation models.65% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/910251API
curl https://kongchang.com/api/v1/knowledge/claims/910251MCP
get_claim(id=910251)