LongCat-2.0 model card (Hugging Face) Jul 5, 2026
- Terminal-Bench 2.1 70.8
- SWE-bench Pro 59.5
- SWE-bench Multilingual 77.3
- FORTE 73.2
- BrowseComp 79.9
- RWSearch 78.8
- IFEval 90
- Writing Bench 83.8
- IMO-AnswerBench 81.8
- GPQA-diamond 88.9
LongCat-2.0
LongCat-2.0 ranks first on StudyArena Text.
14
Benchmarks with a score
1
Ranked first, out of five or more models
1
In the top three
Meituan Jul 2026 LongCat-2.0
Open weights Flagship
152.4
#79 of 404
Self-Reported Index, from the lab's own numbers. No Epoch Capabilities Index score yet.
- FORTE 73.2% #7 /21
- ClawProBench 58.8% #25 /49
- Claw-SWE-Bench 67.1% #12 /16
Where it ranks
Its best score on each benchmark, and where that ranks among every model with a score there. We rank each model once, at its best setting.
Knowledge and language
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| StudyArena Text | 1,375 | #1 of 85 | Official leaderboard | |
| IFEval | 90% | #30 of 135 | Lab model card |
Math and science
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| GPQA Diamond | 88.9% | #46 of 302 | Lab model card | |
| IMO-AnswerBench | 81.8% | #22 of 37 | Lab model card |
Coding
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| SWE-bench Multilingual | 77.3% | #24 of 73 | Lab model card | |
| SWE-Bench Pro | 59.5% | #28 of 81 | Lab model card | |
| ALE-Bench | 1,212 | #81 of 126 | Official leaderboard | |
| Claw-SWE-Bench | 67.1% | #12 of 16 | Official leaderboard |
Agents and tools
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| BrowseComp | 79.9% | #28 of 81 | Lab model card | |
| ClawProBench | 58.8% | #25 of 49 | Official leaderboard | |
| Terminal-Bench 2.1 | 70.8% | #38 of 56 | Lab model card | |
| RWSearch | 78.8% | #2 of 2 | Lab model card |
Professional work
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| FORTE | 73.2% | #7 of 21 | Official leaderboard |
Safety and security
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| nocot-bench (NCRI) | 82.5 | #52 of 185 | Official leaderboard |
Its lab's numbers put it at 152.4, number 79 of 404
The Self-Reported Index fits one score to the 7 capability results Meituan reported for this model, on the scale of the Epoch Capabilities Index. Labs choose which benchmarks to show and how to run them, so read it as the lab's claim, not an independent measurement. The 90% interval runs from 149.6 to 155.2. How we compute it.
- IFEval 90
- GPQA Diamond 88.9
- IMO-AnswerBench 81.8
- BrowseComp 79.9
- SWE-bench Multilingual 77.3
- Terminal-Bench 2.1 70.8
- SWE-Bench Pro 59.5
Its lab's number next to a leaderboard
Where the benchmark's own leaderboard, or Epoch AI's run, also has this model. Gaps mostly come from different harnesses, effort levels and task subsets. Every pair.
Its launch post
The benchmarks the lab chose to headline, with the scores it gave.