AI models
LongCat-Flash
Training compute
3.7×10²⁴ FLOP
Parameters
560B
Published
Sep 1, 2025
LongCat-Flash is an AI model developed by Meituan Inc (China), first published in September 2025. It works in the language domain, on tasks such as language modeling/generation, question answering, chat and 3 more.
Training it took an estimated 3.7×10²⁴ FLOP of compute (estimation method: operation counting). The model has 560,000,000,000 parameters. It was trained on roughly 23T datapoints.
Access: Open weights (unrestricted). Its weights are openly available. Epoch AI rates the confidence of this record as likely.
Full record
- Organization
- Meituan Inc
- Country of organization
- China
- Domain
- Language
- Task
- Language modeling/generation, Question answering, Chat, Code generation, Quantitative reasoning, Instruction interpretation
- Training compute
- 3.7×10²⁴ FLOP
- Compute estimation method
- Operation counting
- Parameters
- 560,000,000,000
- Dataset size
- 23T
- Training time
- 720 h
- Model accessibility
- Open weights (unrestricted)
- Open weights
- Yes
- Epoch confidence
- Likely
SourceEpoch AI, 'AI Models'. Published online at epoch.ai. Retrieved 2026-07-29 from https://epoch.ai/data/ai-models. Licensed under CC BY 4.0.
← All ai models