Model comparison
GPT-6.1 Sol vs Nemotron 3 Ultra
GPT-6.1 Sol is the stronger model overall, scoring 65.6 to 42.5 on the Noometry Index. Nemotron 3 Ultra costs 4.3× less per token, which makes it the better buy when GPT-6.1 Sol's lead doesn't matter for your workload.
Last verified . 22 shared benchmarks.
NVIDIASummary
- They share 22 benchmarks with published results for both. GPT-6.1 Sol scores higher in 8 categories and Nemotron 3 Ultra in 1 category; 7 gaps are clear of the uncertainty.
- The widest gap is in math, where GPT-6.1 Sol leads 93.7 to 35.3.
- The biggest single-benchmark swing is ProofBench: 99% for GPT-6.1 Sol and 2% for Nemotron 3 Ultra.
- Nemotron 3 Ultra is cheaper at $0.50 / $2.20 per million input/output tokens, against $2 / $10 for GPT-6.1 Sol.
- GPT-6.1 Sol accepts more context: 1.05M tokens versus 262K.
- Nemotron 3 Ultra has downloadable open weights; the other is API-only.
Side by side
| GPT-6.1 Sol | Nemotron 3 Ultra | |
|---|---|---|
| Provider | OpenAI | NVIDIA |
| Noometry Index | 65.6 | 42.5 |
| Released | 2026-09-29 | 2026-06-04 |
| Weights | Proprietary | Open |
| Context window | 1.05M | 262K |
| Max output | 128K | 128K |
| Input $ / M tokens | $2 | $0.50 |
| Output $ / M tokens | $10 | $2.20 |
| Results tracked | 34 | 30 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-6.1 Sol leads
GPT-6.1 Sol: 63.2 (#8), Nemotron 3 Ultra: 38.1 (#182)
| Benchmark | GPT-6.1 Sol | Nemotron 3 Ultra |
|---|---|---|
| FrontierCode | 50.2% | 13.6% |
| SciCode | 55.8% | 40.3% |
| LMArena Coding | 1487 | 1468 |
| DeepSWE | 75.2% | — |
| LMArena WebDev | 1755 | — |
| WeirdML | — | 43.5% |
Agentic & Tool Use GPT-6.1 Sol leads
GPT-6.1 Sol: 39.6 (#26), Nemotron 3 Ultra: 23.4 (#126)
| Benchmark | GPT-6.1 Sol | Nemotron 3 Ultra |
|---|---|---|
| APEX-Agents | 60% | 22.7% |
| GDP.pdf | 32% | — |
Reasoning GPT-6.1 Sol leads
GPT-6.1 Sol: 81.9 (#2), Nemotron 3 Ultra: 31.0 (#82)
| Benchmark | GPT-6.1 Sol | Nemotron 3 Ultra |
|---|---|---|
| CritPt | 31.7% | 3.1% |
| Chess Puzzles | 61% | 12% |
| LMArena Hard Prompts | 1466 | 1452 |
| Mystery Game Puzzles | 80% | 20% |
| Epoch Capabilities Index | 166.09 | 146.17 |
| ARC-AGI-2 | 94.2% | — |
| NYT Connections (extended) | 95.5% | — |
| ARC-AGI-1 | 98.5% | — |
| EBR-Bench | 54.3% | — |
| DTBench | — | 90.1% |
| LMCA | — | 36.9% |
Math GPT-6.1 Sol leads
GPT-6.1 Sol: 93.7 (#1), Nemotron 3 Ultra: 35.3 (#187)
| Benchmark | GPT-6.1 Sol | Nemotron 3 Ultra |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 100% | 86.7% |
| ProofBench | 99% | 2% |
| LMArena Math | 1464 | 1457 |
| FrontierMath (Tiers 1-3) | 93.7% | — |
| FrontierMath Tier 4 | 100% | — |
Knowledge GPT-6.1 Sol leads
GPT-6.1 Sol: 71.8 (#4), Nemotron 3 Ultra: 52.5 (#61)
| Benchmark | GPT-6.1 Sol | Nemotron 3 Ultra |
|---|---|---|
| GPQA Diamond | 95.4% | 85.4% |
| LMArena Expert | 1502 | 1472 |
| SimpleQA Verified | 73.9% | — |
Multimodal Not comparable
GPT-6.1 Sol: 52.7 (#5), Nemotron 3 Ultra: —
| Benchmark | GPT-6.1 Sol | Nemotron 3 Ultra |
|---|---|---|
| LMArena Vision | 1288 | — |
| Furniture Assembly | 80% | — |
Multilingual Too close to call
GPT-6.1 Sol: 54.3 (#46), Nemotron 3 Ultra: 53.5 (#64)
| Benchmark | GPT-6.1 Sol | Nemotron 3 Ultra |
|---|---|---|
| LMArena Non-English | 1438 | 1427 |
| LMArena Chinese | 1477 | 1497 |
| LMArena Russian | 1455 | 1417 |
| LMArena French | — | 1472 |
| LMArena German | — | 1471 |
| LMArena Korean | — | 1386 |
| LMArena Spanish | — | 1454 |
Instruction Following GPT-6.1 Sol leads
GPT-6.1 Sol: 77.0 (#29), Nemotron 3 Ultra: 74.7 (#88)
| Benchmark | GPT-6.1 Sol | Nemotron 3 Ultra |
|---|---|---|
| LMArena Instruction Following | 1468 | 1418 |
Long Context Too close to call
GPT-6.1 Sol: 44.9 (#54), Nemotron 3 Ultra: 43.9 (#81)
| Benchmark | GPT-6.1 Sol | Nemotron 3 Ultra |
|---|---|---|
| LMArena Longer Query | 1465 | 1437 |
Writing & Preference Nemotron 3 Ultra leads
GPT-6.1 Sol: 63.6 (#63), Nemotron 3 Ultra: 66.3 (#36)
| Benchmark | GPT-6.1 Sol | Nemotron 3 Ultra |
|---|---|---|
| LMArena Text | 1447 | 1445 |
| LMArena Creative Writing | 1432 | 1398 |
| LMArena Multi-Turn | 1449 | 1411 |
| EQ-Bench Creative Writing | — | 1692 |
Frequently asked questions
Is GPT-6.1 Sol better than Nemotron 3 Ultra?
GPT-6.1 Sol is the stronger model overall, scoring 65.6 to 42.5 on the Noometry Index. Nemotron 3 Ultra costs 4.3× less per token, which makes it the better buy when GPT-6.1 Sol's lead doesn't matter for your workload.
Which is cheaper, GPT-6.1 Sol 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; GPT-6.1 Sol lists at $2 and $10.
Is GPT-6.1 Sol or Nemotron 3 Ultra better for coding?
GPT-6.1 Sol scores higher on coding benchmarks: 63.2 versus 38.1 in the Noometry coding category.
Which has the bigger context window?
GPT-6.1 Sol does, with 1.05M tokens against 262K.
How many benchmarks do GPT-6.1 Sol and Nemotron 3 Ultra share?
22 benchmarks have published results for both models. GPT-6.1 Sol has 34 scored results on Noometry and Nemotron 3 Ultra has 30.