Model comparison
Nemotron 3 Ultra vs Trinity Large Thinking
Nemotron 3 Ultra is the stronger model overall, scoring 42.5 to 38.6 on the Noometry Index. Trinity Large Thinking costs 2.4× less per token, which makes it the better buy when Nemotron 3 Ultra's lead doesn't matter for your workload.
Last verified . 18 shared benchmarks.
NVIDIA
Arcee AISummary
- They share 18 benchmarks with published results for both. Nemotron 3 Ultra scores higher in 7 categories and Trinity Large Thinking in 1 category; 8 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Nemotron 3 Ultra leads 31.0 to 16.9.
- Trinity Large Thinking is cheaper at $0.25 / $0.80 per million input/output tokens, against $0.50 / $2.20 for Nemotron 3 Ultra.
Side by side
| Nemotron 3 Ultra | Trinity Large Thinking | |
|---|---|---|
| Provider | NVIDIA | Arcee AI |
| Noometry Index | 42.5 | 38.6 |
| Released | 2026-06-04 | 2026-04-01 |
| Weights | Open | Open |
| Context window | 262K | 262K |
| Max output | 128K | 80K |
| Input $ / M tokens | $0.50 | $0.25 |
| Output $ / M tokens | $2.20 | $0.80 |
| Results tracked | 30 | 24 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Nemotron 3 Ultra leads
Nemotron 3 Ultra: 38.1 (#182), Trinity Large Thinking: 34.1 (#244)
| Benchmark | Nemotron 3 Ultra | Trinity Large Thinking |
|---|---|---|
| SciCode | 40.3% | 36.1% |
| LMArena Coding | 1468 | 1381 |
| FrontierCode | 13.6% | — |
| LMArena WebDev | — | 1238 |
| WeirdML | 43.5% | — |
Agentic & Tool Use Not comparable
Nemotron 3 Ultra: 23.4 (#126), Trinity Large Thinking: —
| Benchmark | Nemotron 3 Ultra | Trinity Large Thinking |
|---|---|---|
| APEX-Agents | 22.7% | — |
Reasoning Nemotron 3 Ultra leads
Nemotron 3 Ultra: 31.0 (#82), Trinity Large Thinking: 16.9 (#298)
| Benchmark | Nemotron 3 Ultra | Trinity Large Thinking |
|---|---|---|
| CritPt | 3.1% | 0.9% |
| LMArena Hard Prompts | 1452 | 1350 |
| NYT Connections (extended) | — | 16.5% |
| Chess Puzzles | 12% | — |
| Thematic Generalization | — | 41.6% |
| Mystery Game Puzzles | 20% | — |
| DTBench | 90.1% | — |
| LMCA | 36.9% | — |
| Surface Evolver Bench | — | 15.6% |
| Epoch Capabilities Index | 146.17 | — |
Math Trinity Large Thinking leads
Nemotron 3 Ultra: 35.3 (#187), Trinity Large Thinking: 37.6 (#149)
| Benchmark | Nemotron 3 Ultra | Trinity Large Thinking |
|---|---|---|
| LMArena Math | 1457 | 1366 |
| OTIS Mock AIME 2024-2025 | 86.7% | — |
| ProofBench | 2% | — |
Knowledge Nemotron 3 Ultra leads
Nemotron 3 Ultra: 52.5 (#61), Trinity Large Thinking: 40.9 (#113)
| Benchmark | Nemotron 3 Ultra | Trinity Large Thinking |
|---|---|---|
| LMArena Expert | 1472 | 1360 |
| GPQA Diamond | 85.4% | — |
| Vectara Hallucination Rate | — | 6.9% |
Multilingual Nemotron 3 Ultra leads
Nemotron 3 Ultra: 53.5 (#64), Trinity Large Thinking: 46.2 (#160)
| Benchmark | Nemotron 3 Ultra | Trinity Large Thinking |
|---|---|---|
| LMArena Non-English | 1427 | 1325 |
| LMArena Chinese | 1497 | 1373 |
| LMArena French | 1472 | 1374 |
| LMArena German | 1471 | 1356 |
| LMArena Korean | 1386 | 1306 |
| LMArena Russian | 1417 | 1337 |
| LMArena Spanish | 1454 | 1357 |
| LMArena Japanese | — | 1311 |
Instruction Following Nemotron 3 Ultra leads
Nemotron 3 Ultra: 74.7 (#88), Trinity Large Thinking: 70.5 (#162)
| Benchmark | Nemotron 3 Ultra | Trinity Large Thinking |
|---|---|---|
| LMArena Instruction Following | 1418 | 1334 |
Long Context Nemotron 3 Ultra leads
Nemotron 3 Ultra: 43.9 (#81), Trinity Large Thinking: 41.3 (#144)
| Benchmark | Nemotron 3 Ultra | Trinity Large Thinking |
|---|---|---|
| LMArena Longer Query | 1437 | 1355 |
Writing & Preference Nemotron 3 Ultra leads
Nemotron 3 Ultra: 66.3 (#36), Trinity Large Thinking: 53.8 (#158)
| Benchmark | Nemotron 3 Ultra | Trinity Large Thinking |
|---|---|---|
| LMArena Text | 1445 | 1340 |
| LMArena Creative Writing | 1398 | 1320 |
| LMArena Multi-Turn | 1411 | 1342 |
| EQ-Bench Creative Writing | 1692 | — |
Frequently asked questions
Is Nemotron 3 Ultra better than Trinity Large Thinking?
Nemotron 3 Ultra is the stronger model overall, scoring 42.5 to 38.6 on the Noometry Index. Trinity Large Thinking costs 2.4× less per token, which makes it the better buy when Nemotron 3 Ultra's lead doesn't matter for your workload.
Which is cheaper, Nemotron 3 Ultra or Trinity Large Thinking?
Trinity Large Thinking is cheaper. It lists at $0.25 per million input tokens and $0.80 per million output tokens; Nemotron 3 Ultra lists at $0.50 and $2.20.
Is Nemotron 3 Ultra or Trinity Large Thinking better for coding?
Nemotron 3 Ultra scores higher on coding benchmarks: 38.1 versus 34.1 in the Noometry coding category.
Which has the bigger context window?
Both accept 262K tokens.
How many benchmarks do Nemotron 3 Ultra and Trinity Large Thinking share?
18 benchmarks have published results for both models. Nemotron 3 Ultra has 30 scored results on Noometry and Trinity Large Thinking has 24.