- Best of AIME-25 (Avg@16) w/ tools 99.6 / Heavy Thinking 100.0 100
- Best of HMMT-25 (Avg@16) w/ tools 93.4 / Heavy Thinking 97.5 97.5
- Best of IMO-AnswerBench (Avg@4) 78.6 / Heavy Thinking 86.8 86.8
- Best of AMO-Bench EN (Avg@16) 61.6 / Heavy Thinking 66.0 66
- Best of BrowseComp (Pass@1) 56.6 / with context management 73.1 73.1
- Best of BrowseComp-zh (Pass@1, 24 answers revised) 69.0 / with context management 77.7 77.7
- RW Search (Pass@1) 79.5
- τ²-Avg (Avg@4): Retail 88.6 / Airline 76.5 / Telecom 99.3 88.2
- τ²-Noise (Avg@4) 67.1
- VitaBench (Avg@4) 29.3
- VitaBench-Noise (Avg@4) 20.5
- Random Complex Tasks (Avg@4) 35.8
- HLE text-only (w/o tools) 25.2
- Best of GPQA-Diamond (Avg@16) 80.5 / Heavy Thinking 85.2 85.2
- LCB (24.08-25.05) (Avg@4) 82.8
- OJBench (Pass@1) 42.2
- OIBench EN (Pass@1) 47.7
- SWE-bench Verified (Avg@5) 70
LongCat-Flash-Thinking-2601
LongCat-Flash-Thinking-2601 places in the top three on 3 benchmarks.
19
Benchmarks with a score
0
Ranked first, out of five or more models
3
In the top three
Meituan Jan 2026 LongCat-Flash-Thinking-2601
Open weights Flagship
No Epoch Capabilities Index score yet
- VITA-Bench 29.3% #3 /18
- BrowseComp-ZH 77.7% #3 /17
- RWSearch 79.5% #2 /7
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 | |
|---|---|---|---|---|
| Humanity's Last Exam | 25.2% | #87 of 147 | Model card |
Math and science
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| AIME 2025 | 100% | #8 of 143 | Model card | |
| HMMT 2025 | 97.5% | #7 of 49 | Model card | |
| IMO-AnswerBench | 86.8% | #9 of 37 | Model card | |
| GPQA Diamond | 85.2% | #71 of 284 | Model card | |
| AMO-Bench | 66% | #1 of 4 | Model card | |
| General365 | 50.8% | #16 of 23 | Official leaderboard |
Coding
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| LiveCodeBench v6 | 82.8% | #25 of 76 | Model card | |
| OJBench | 42.2% | #5 of 12 | Model card | |
| SWE-bench Verified | 70% | #77 of 150 | Model card | |
| OIBench | 47.7% | #1 of 1 | Model card |
Agents and tools
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| VITA-Bench | 29.3% | #3 of 18 | Official leaderboard | |
| BrowseComp-ZH | 77.7% | #3 of 17 | Model card | |
| tau2-bench | 88.2% | #17 of 69 | Model card | |
| RWSearch | 79.5% | #2 of 7 | Official leaderboard | |
| BrowseComp | 73.1% | #43 of 77 | Model card | |
| Random Complex Tasks (LongCat) | 35.8% | #1 of 1 | Model card | |
| tau2-Noise (LongCat) | 67.1% | #1 of 1 | Model card | |
| VitaBench-Noise (LongCat) | 20.5% | #1 of 1 | Model card |
Its launch post
The benchmarks the lab chose to headline, with the scores it gave.
Its system card
Every result in the longer documents the lab published with it.
LongCat-Flash-Thinking-2601 Technical Report 23 Jan 2026
Capabilities
- AIME 2025 100
- HMMT 2025 97.5
- IMO-AnswerBench 86.8
- AMO-Bench 66
- BrowseComp 73.1
- BrowseComp-ZH 77.7
- RWSearch 79.5
- tau2-bench 88.2
- VITA-Bench 29.3
- Random Complex Tasks (LongCat) 35.8
- Humanity's Last Exam 25.2
- GPQA Diamond 85.2
- LiveCodeBench v6 82.8
- OJBench 42.2
- OIBench 47.7
- SWE-bench Verified 70