LongCat-2.0 Publisher-Reported Benchmark Scores
LongCat-2.0 Publisher-Reported Scores
The following scores were published by the LongCat team in the LongCat-2.0 release blog post (June 30, 2026) and the model card in the LongCat-2.0 GitHub repository. These are publisher-reported and have not been independently verified. See the methodology page for the full evidence labelling policy.
| Benchmark | Score | Description |
|---|---|---|
| SWE-bench Pro | 59.5 | Software engineering tasks on real repositories |
| SWE-bench Multilingual | 77.3 | Multi-language software engineering |
| Terminal-Bench 2.1 | 70.8 | Real terminal command execution and error recovery |
| BrowseComp | 79.9 | Complex browsing and information retrieval |
| RWSearch | 78.8 | Search agent evaluation |
| FORTE | 73.2 | Productivity and office scenario tasks |
Source: LongCat-2.0 release blog post (longcat.chat/blog/longcat-2.0), published 2026-06-30. All scores labelled publisher-reported.
Limitations
All scores on this page are publisher-reported, not independently tested. Source attribution is included for each data point. No independent verification of these results has been performed. Publishers may select benchmarks, configurations, and baselines that present their model favorably.
Sources
- LongCat-2.0 blog post
Publisher documentationPublished 2026-06-30Accessed 2026-07-24
The publisher's official LongCat-2.0 release post containing vendor-reported benchmark scores across coding, agent, and search benchmarks. All scores on this benchmark page are sourced from this post unless otherwise noted.
- LongCat-2.0 GitHub Repository model card
Primary sourceAccessed 2026-07-24
The README in this repository includes a benchmark results table with scoring labels indicating which results are publisher-reported versus sourced from third-party reports.
Independent third-party disclosure
This page is published by an independent third-party site. It is not affiliated with, endorsed by, sponsored by, or operated by Meituan, LongCat, or any of their affiliates. The content summarizes publicly-available primary documentation and does not represent the views of any referenced organization.
Last reviewed: 2026-06-30