Introducing Mercury 2.5 (Inception blog) 8 Sep 2026
Mercury 2.5
Mercury 2.5 ranks first on tau3-bench.
Inception Labs Released 8 Sep 2026
24
Benchmarks with a score
1
Ranked first, out of five or more models
1
In the top three
Inception Labs Sep 2026
Mercury 2.5
Closed weights
No Epoch Capabilities Index score yet
- IFBench 77% #16 /70
- SciCode 38% #35 /47
- AA-LCR (Artificial Analysis Long Context Reasoning) 68% #12 /31
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 | |
|---|---|---|---|---|
| IFBench | 77% | #16 of 70 | Model card | |
| AA-LCR (Artificial Analysis Long Context Reasoning) | 68% | #12 of 31 | Model card | |
| AA-Omniscience | 22% | #11 of 17 | Model card |
Math and science
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| GPQA Diamond | 79% | #118 of 284 | Model card | |
| ProofBench | 3.00% | #79 of 83 | Epoch AI | |
| ProofBench v1.1 | 3.00% | #49 of 51 | Official leaderboard | |
| CritPt | 0% | #160 of 160 | Epoch AI |
Coding
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| SciCode | 38% | #35 of 47 | Model card | |
| Vibe Code Bench | 3.68% | #103 of 115 | Official leaderboard | |
| ALE-Bench | 302 | #113 of 121 | Epoch AI | |
| Code Migration | 4.52% | #74 of 78 | Official leaderboard | |
| IOI v2 (Vals) | 2.44% | #45 of 45 | Official leaderboard |
Agents and tools
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| tau3-bench | 96% | #1 of 12 | Model card | |
| DeepSearchQA | 34% | #22 of 22 | Model card |
Professional work
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| LegalBench | 83.1% | #61 of 138 | Official leaderboard | |
| MedCode | 31.3% | #88 of 96 | Official leaderboard | |
| Public Benefits Bench | 45.3% | #47 of 50 | Official leaderboard | |
| Legal Research Bench | 4.33% | #83 of 88 | Official leaderboard | |
| Vals Finance Agent v2 | 18.9% | #77 of 80 | Official leaderboard | |
| MedScribe | 55.1% | #95 of 98 | Official leaderboard | |
| Excel Modeling Benchmark | 8.79% | #75 of 76 | Official leaderboard | |
| Harvey LAB (Legal Agent Benchmark) | 0% | #80 of 80 | Official leaderboard | |
| Tax Agent Bench | 2.49% | #79 of 79 | Official leaderboard |
Safety and security
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| CyberBench v1.1 (Vals) | 45.5% | #36 of 43 | Official leaderboard |
Its launch post
The benchmarks the lab chose to headline, with the scores it gave.