- GPQA-Diamond 89.5
- HLE, text-only (no tools); HLE full set 29.20 31.8
- MMLU-ProX 76.9
- AA-Omniscience index 4.03 (accuracy 26.20, hallucination rate 30.0) 4.03
- CritPt 10.9
- APEX shortlist 88.4
- LiveCodeBench v6 88.2
- SciCode, subtask 47.1
- AA-LCR v1.1 68.6
- GraphWalks 89.7
- IFBench 70.9
- LMArena v3 proxy (NV Chat Benchmark in the technical report) 52.7
- WMT24++ (COMET) 84.8
- Mean of tau2-bench Retail 69.24 / Airline 76.50 / Telecom 73.19 73
- tau3-bench banking 18.9
- GDPval-AA v2 1211
- BrowseComp (78.91 in the technical report) 78
- SWE-bench Verified (OpenCode harness) 71.1
- SWE-bench Multilingual (OpenCode harness) 67.7
- SWE-bench Pro (OpenCode harness) 51.8
- Terminal-Bench 2.1 (Terminus 2) 60.7
- BabyVision 29.6
- MMLongBench-Doc 62.8
- V* 85.6
- OCR-Reasoning 65.1
- MathVision 78.2
- OCRBench v2, EN 68.12 / ZH 59.15 68.1
- CharXiv reasoning 80.8
- Video-MME v2 (subtitle setting not stated) 32.9
- OSWorld-Verified 51.1
- WebVoyager [offline] 77.7
- AutomationBench-AA [offline] 32.4
- GDP.pdf [offline] 6.6 (AA 6.6) 6.6
- Terminal Bench 4 [offline] 1
- DeepSWE [offline] 24.6
Nemotron 3.5 Super VL
Nemotron 3.5 Super VL ranks first on MMLongBench-Doc.
55
Benchmarks with a score
1
Ranked first, out of five or more models
4
In the top three
NVIDIA Oct 2026 Nemotron 3.5 Super VL
Open weights Flagship
No Epoch Capabilities Index score yet
- IFBench 70.9% #41 /70
- SciCode 47.1% #16 /47
- AA-Omniscience 4.03% #17 /17
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 | |
|---|---|---|---|---|
| GraphWalks | 89.7% | #6 of 21 | Model card | |
| LMArena (Elo) | 1,412 | #115 of 333 | Official leaderboard | |
| Humanity's Last Exam | 31.8% | #71 of 147 | Model card | |
| WMT24++ | 84.8% | #6 of 12 | Model card | |
| NV Chat Benchmark | 52.7 | #1 of 2 | Model card | |
| IFBench | 70.9% | #41 of 70 | Model card | |
| MMLU-ProX | 76.9% | #12 of 17 | Model card | |
| AA-LCR 1.1 | 68.6% | #4 of 5 | Model card | |
| AA-Omniscience | 4.03% | #17 of 17 | Model card |
Math and science
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| GPQA Diamond | 89.5% | #33 of 284 | Model card | |
| MathArena (Apex, IMO, USAMO) | 88.4% | #2 of 13 | Model card | |
| CritPt | 10.9% | #10 of 14 | Model card | |
| ARC-AGI-2 | 1.94% | #75 of 95 | Official leaderboard | |
| ARC-AGI | 22.7% | #76 of 92 | Official leaderboard |
Coding
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| LiveCodeBench v6 | 88.2% | #10 of 76 | Model card | |
| Bird-SQL | 59.6% | #2 of 13 | Model card | |
| SciCode | 47.1% | #16 of 47 | Model card | |
| SWE-bench Verified | 71.1% | #68 of 150 | Model card | |
| SWE-bench Multilingual | 67.7% | #44 of 70 | Model card | |
| Code Arena | 1,399 | #82 of 124 | Official leaderboard | |
| SWE-Bench Pro | 51.8% | #60 of 77 | Model card | |
| Vibe Code Bench | 3.54% | #104 of 115 | Official leaderboard | |
| Image-to-WebDev Arena | 1,407 | #51 of 55 | Official leaderboard | |
| Code Migration v1.1 | 16.4% | #27 of 29 | Official leaderboard | |
| DeepSWE | 24.6% | #51 of 51 | Model card |
Agents and tools
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| BrowseComp | 78.9% | #29 of 77 | Model card | |
| AutomationBench | 32.4% | #23 of 41 | Model card | |
| tau2-bench | 73% | #47 of 69 | Model card | |
| τ³-bench Banking | 18.9% | #14 of 20 | Model card | |
| Terminal-Bench 2.1 | 60.7% | #42 of 53 | Model card | |
| WebVoyager | 77.7% | #9 of 11 | Model card | |
| OSWorld-Verified | 52.1% | #33 of 37 | Model card | |
| Agent Arena | -0.15 | #49 of 53 | Official leaderboard | |
| ScreenSpot-Pro | 43.5% | #24 of 25 | Model card | |
| Terminal-Bench 4.0 | 1.00% | #23 of 23 | Model card |
Professional work
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| Tax Agent Bench | 11.6% | #62 of 79 | Official leaderboard | |
| GDPval-AA v2 | 1,211 | #22 of 26 | Model card | |
| Legal Research Bench | 10.1% | #79 of 88 | Official leaderboard | |
| Vals Index v2.2 | 29.8% | #30 of 32 | Official leaderboard | |
| Harvey LAB v1.1 (HLAB-Vals-1.1) | 3.75% | #31 of 33 | Official leaderboard | |
| Vals Finance Index | 41% | #33 of 35 | Official leaderboard | |
| Excel Modeling Benchmark v1.1 | 52.8% | #35 of 37 | Official leaderboard | |
| Vals Legal Index | 55.4% | #32 of 33 | Official leaderboard | |
| Vals Finance Agent v2.1 | 47.7% | #34 of 35 | Official leaderboard |
Vision, audio and robotics
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| OCRBench v2 | 67.4% | #2 of 65 | Official leaderboard | |
| MMLongBench-Doc | 62.8% | #1 of 13 | Model card | |
| OCR-Reasoning | 65.1% | #1 of 3 | Model card | |
| MATH-Vision (MathVision) | 78.2% | #18 of 42 | Model card | |
| CharXiv | 80.8% | #29 of 67 | Model card | |
| Vision Arena | 1,204 | #76 of 136 | Official leaderboard | |
| V* | 85.6% | #9 of 12 | Model card | |
| BabyVision | 29.6% | #19 of 20 | Model card | |
| GDP.pdf | 6.60% | #15 of 15 | Model card | |
| VideoMME v2 (with subtitles) | 32.9% | #6 of 6 | Model card |
Other
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| LiveBench | 50.2% | #26 of 28 | Official leaderboard |
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. 3 of them aren't in the launch post.
Capabilities
- SWE-bench Verified 71.1
- SWE-bench Multilingual 67.7
- SWE-Bench Pro 51.8
- tau2-bench
- τ³-bench Banking 18.9
- Terminal-Bench 2.1 60.7
- GDPval-AA v2 1211.2
- BrowseComp 78.9
- SciCode 47.1
- LiveCodeBench v6 88.2
- GPQA Diamond 89.5
- CritPt 10.9
- Humanity's Last Exam 31.8
- MathArena (Apex, IMO, USAMO) 88.4
- BabyVision 29.6
- CharXiv 80.8
- MATH-Vision (MathVision) 78.2
- OCR-Reasoning 65.1
- OCRBench v2 68.1
- VideoMME v2 (with subtitles) 32.9
- V* 85.6
- MMLongBench-Doc 62.8
- OSWorld-Verified 52.1
- AA-LCR 1.1 68.6
- GraphWalks 89.7
- IFBench 70.9
- NV Chat Benchmark 52.7
- MMLU-ProX 76.9
- WMT24++ 84.8
- Bird-SQL 59.6
- ScreenSpot-Pro 43.5
- LMArena (Elo) 1420