Model comparison
Claude Opus 4.8 vs Nemotron 3 Ultra
Claude Opus 4.8 is the stronger model overall, scoring 60.7 to 42.5 on the Noometry Index. Nemotron 3 Ultra costs 11× less per token, which makes it the better buy when Claude Opus 4.8's lead doesn't matter for your workload.
Last verified . 30 shared benchmarks.
NVIDIASummary
- They share 30 benchmarks with published results for both. Claude Opus 4.8 scores higher in 9 categories and Nemotron 3 Ultra in 0 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in math, where Claude Opus 4.8 leads 78.4 to 35.3.
- The biggest single-benchmark swing is ProofBench: 69% for Claude Opus 4.8 and 2% for Nemotron 3 Ultra.
- Nemotron 3 Ultra is cheaper at $0.50 / $2.20 per million input/output tokens, against $5 / $25 for Claude Opus 4.8.
- Claude Opus 4.8 accepts more context: 1M tokens versus 262K.
- Nemotron 3 Ultra has downloadable open weights; the other is API-only.
Side by side
| Claude Opus 4.8 | Nemotron 3 Ultra | |
|---|---|---|
| Provider | Anthropic | NVIDIA |
| Noometry Index | 60.7 | 42.5 |
| Released | 2026-05-28 | 2026-06-04 |
| Weights | Proprietary | Open |
| Context window | 1M | 262K |
| Max output | 128K | 128K |
| Input $ / M tokens | $5 | $0.50 |
| Output $ / M tokens | $25 | $2.20 |
| Results tracked | 65 | 30 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Claude Opus 4.8 leads
Claude Opus 4.8: 59.9 (#12), Nemotron 3 Ultra: 38.1 (#182)
| Benchmark | Claude Opus 4.8 | Nemotron 3 Ultra |
|---|---|---|
| FrontierCode | 46.5% | 13.6% |
| SciCode | 53.5% | 40.3% |
| WeirdML | 82.9% | 43.5% |
| LMArena Coding | 1490 | 1468 |
| DeepSWE | 59% | — |
| LMArena WebDev | 1556 | — |
| GSO | 47.1% | — |
| ALE-Bench | 1,564 | — |
Agentic & Tool Use Claude Opus 4.8 leads
Claude Opus 4.8: 47.6 (#11), Nemotron 3 Ultra: 23.4 (#126)
| Benchmark | Claude Opus 4.8 | Nemotron 3 Ultra |
|---|---|---|
| APEX-Agents | 48.9% | 22.7% |
| OSWorld 2.0 | 20.6% | — |
| Remote Labor Index | 8.3% | — |
| τ²-bench Banking | 39.7% | — |
| DeepResearch Bench | 50.2% | — |
| PostTrainBench | 33.8% | — |
| GBAEval | 70.9% | — |
| GDP.pdf | 24% | — |
| LMArena Search | 1204 | — |
| Vending-Bench 2 | 5,787 | — |
Reasoning Claude Opus 4.8 leads
Claude Opus 4.8: 64.7 (#16), Nemotron 3 Ultra: 31.0 (#82)
| Benchmark | Claude Opus 4.8 | Nemotron 3 Ultra |
|---|---|---|
| CritPt | 20.9% | 3.1% |
| Chess Puzzles | 34% | 12% |
| LMArena Hard Prompts | 1482 | 1452 |
| Mystery Game Puzzles | 36% | 20% |
| DTBench | 94.9% | 90.1% |
| LMCA | 57.5% | 36.9% |
| Epoch Capabilities Index | 158.21 | 146.17 |
| ARC-AGI-2 | 72.1% | — |
| SimpleBench | 64.8% | — |
| Kagi LLM Benchmark | 88.8% | — |
| NYT Connections (extended) | 91.1% | — |
| ARC-AGI-1 | 92.5% | — |
| EnigmaEval | 23.5% | — |
| EBR-Bench | 28.6% | — |
| Surface Evolver Bench | 87.5% | — |
| Bench to the Future 3 | 0.14 | — |
| ForecastBench | 59.9 | — |
Math Claude Opus 4.8 leads
Claude Opus 4.8: 78.4 (#13), Nemotron 3 Ultra: 35.3 (#187)
| Benchmark | Claude Opus 4.8 | Nemotron 3 Ultra |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 98.3% | 86.7% |
| ProofBench | 69% | 2% |
| LMArena Math | 1487 | 1457 |
| FrontierMath (Tiers 1-3) | 80% | — |
| FrontierMath Tier 4 | 56.1% | — |
| MathArena Final-Answer Competitions | 91.8% | — |
| FrontierMath (Feb 2025 set) | 47.2% | — |
| FrontierMath Tier 4 (v1) | 31.3% | — |
Knowledge Claude Opus 4.8 leads
Claude Opus 4.8: 61.3 (#29), Nemotron 3 Ultra: 52.5 (#61)
| Benchmark | Claude Opus 4.8 | Nemotron 3 Ultra |
|---|---|---|
| GPQA Diamond | 91% | 85.4% |
| LMArena Expert | 1502 | 1472 |
| SimpleQA Verified | 53% | — |
Multimodal Not comparable
Claude Opus 4.8: 42.9 (#26), Nemotron 3 Ultra: —
| Benchmark | Claude Opus 4.8 | Nemotron 3 Ultra |
|---|---|---|
| LMArena Vision | 1294 | — |
| Blueprint-Bench 2 | 14.5% | — |
| Furniture Assembly | 42.5% | — |
| LMArena Document | 1475 | — |
Multilingual Claude Opus 4.8 leads
Claude Opus 4.8: 55.2 (#33), Nemotron 3 Ultra: 53.5 (#64)
| Benchmark | Claude Opus 4.8 | Nemotron 3 Ultra |
|---|---|---|
| LMArena Non-English | 1450 | 1427 |
| LMArena Chinese | 1507 | 1497 |
| LMArena French | 1481 | 1472 |
| LMArena German | 1472 | 1471 |
| LMArena Korean | 1432 | 1386 |
| LMArena Russian | 1474 | 1417 |
| LMArena Spanish | 1466 | 1454 |
| LMArena Japanese | 1440 | — |
Instruction Following Claude Opus 4.8 leads
Claude Opus 4.8: 77.4 (#24), Nemotron 3 Ultra: 74.7 (#88)
| Benchmark | Claude Opus 4.8 | Nemotron 3 Ultra |
|---|---|---|
| LMArena Instruction Following | 1476 | 1418 |
Long Context Claude Opus 4.8 leads
Claude Opus 4.8: 45.4 (#35), Nemotron 3 Ultra: 43.9 (#81)
| Benchmark | Claude Opus 4.8 | Nemotron 3 Ultra |
|---|---|---|
| LMArena Longer Query | 1483 | 1437 |
Writing & Preference Claude Opus 4.8 leads
Claude Opus 4.8: 72.0 (#16), Nemotron 3 Ultra: 66.3 (#36)
| Benchmark | Claude Opus 4.8 | Nemotron 3 Ultra |
|---|---|---|
| LMArena Text | 1461 | 1445 |
| LMArena Creative Writing | 1454 | 1398 |
| EQ-Bench Creative Writing | 1840 | 1692 |
| LMArena Multi-Turn | 1476 | 1411 |
| EQ-Bench 4 | 1281 | — |
Frequently asked questions
Is Claude Opus 4.8 better than Nemotron 3 Ultra?
Claude Opus 4.8 is the stronger model overall, scoring 60.7 to 42.5 on the Noometry Index. Nemotron 3 Ultra costs 11× less per token, which makes it the better buy when Claude Opus 4.8's lead doesn't matter for your workload.
Which is cheaper, Claude Opus 4.8 or Nemotron 3 Ultra?
Nemotron 3 Ultra is cheaper. It lists at $0.50 per million input tokens and $2.20 per million output tokens; Claude Opus 4.8 lists at $5 and $25.
Is Claude Opus 4.8 or Nemotron 3 Ultra better for coding?
Claude Opus 4.8 scores higher on coding benchmarks: 59.9 versus 38.1 in the Noometry coding category.
Which has the bigger context window?
Claude Opus 4.8 does, with 1M tokens against 262K.
How many benchmarks do Claude Opus 4.8 and Nemotron 3 Ultra share?
30 benchmarks have published results for both models. Claude Opus 4.8 has 65 scored results on Noometry and Nemotron 3 Ultra has 30.