- Terminal-Bench v2.1 (Artificial Analysis score quoted by Upstage) 57
- BrowseComp* (in-house) 49.2
- SWE-Bench Verified (OpenHands)* (in-house) 70.6
- MCP-Atlas* (in-house) 61.4
- APEX-Agents* (in-house) 18.7
- τ³-Banking (Artificial Analysis score quoted by Upstage) 23
- AA-LCR (Artificial Analysis score quoted by Upstage) 71
- GPQA Diamond (Artificial Analysis score quoted by Upstage) 89
- MMLU-Pro* (in-house) 86.3
- AIME 2026* (in-house) 95.3
- KMMLU-Pro* (in-house) 79.2
- KBL (law)* (in-house) 77.5
- KorMedMCQA (medicine)* (in-house) 93.2
- Ko-GDPval* (in-house) 87.3
Solar Pro 4
Solar Pro 4 ranks first on KMMLU-Pro.
17
Benchmarks with a score
1
Ranked first, out of five or more models
1
In the top three
Upstage Aug 2026 Solar Pro 4
Closed weights Flagship
No Epoch Capabilities Index score yet
- AA-LCR (Artificial Analysis Long Context Reasoning) 71% #7 /31
- KMMLU-Pro 79.2% #1 /7
- BrowseComp 49.2% #60 /77
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 | |
|---|---|---|---|---|
| MMLU-Pro | 86.3% | #15 of 151 | Model card | |
| KMMLU-Pro | 79.2% | #1 of 7 | Model card | |
| AA-LCR (Artificial Analysis Long Context Reasoning) | 71% | #7 of 31 | Model card | |
| LMArena (Elo) | 1,388 | #144 of 333 | Official leaderboard |
Math and science
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| GPQA Diamond | 89% | #39 of 284 | Model card | |
| AIME 2026 | 95.3% | #12 of 34 | Model card |
Coding
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| SWE-bench Verified | 70.6% | #73 of 150 | Model card | |
| Code Arena | 1,370 | #90 of 124 | Official leaderboard |
Agents and tools
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| τ³-bench Banking | 23% | #11 of 20 | Model card | |
| BrowseComp | 49.2% | #60 of 77 | Model card | |
| MCP Atlas | 61.4% | #32 of 40 | Model card | |
| Terminal-Bench 2.1 | 57% | #44 of 53 | Model card | |
| Agent Arena | -0.21 | #53 of 53 | Official leaderboard |
Professional work
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| KBL | 77.5% | #1 of 4 | Model card | |
| KorMedMCQA | 93.2% | #1 of 4 | Model card | |
| Ko-GDPval | 87.3% | #1 of 2 | Model card | |
| APEX-Agents | 18.7% | #11 of 12 | Model card |
Its launch post
The benchmarks the lab chose to headline, with the scores it gave.