Model comparison
Inkling vs Nemotron 3 Ultra
Inkling is the stronger model overall, scoring 44.1 to 42.5 on the Noometry Index. Nemotron 3 Ultra costs 2.5× less per token, which makes it the better buy when Inkling's lead doesn't matter for your workload.
Last verified . 29 shared benchmarks.
NVIDIASummary
- They share 29 benchmarks with published results for both. Inkling scores higher in 5 categories and Nemotron 3 Ultra in 4 categories; 6 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Inkling leads 40.4 to 31.0.
- The biggest single-benchmark swing is WeirdML: 32.3% for Inkling and 43.5% for Nemotron 3 Ultra.
- Nemotron 3 Ultra is cheaper at $0.60 / $2.40 per million input/output tokens, against $1.87 / $4.68 for Inkling.
- Nemotron 3 Ultra accepts more context: 262K tokens versus 66K.
Side by side
| Inkling | Nemotron 3 Ultra | |
|---|---|---|
| Provider | Thinking Machines Lab | NVIDIA |
| Noometry Index | 44.1 | 42.5 |
| Released | 2026-07-15 | 2026-06-04 |
| Weights | Open | Open |
| Context window | 66K | 262K |
| Max output | 66K | 128K |
| Input $ / M tokens | $1.87 | $0.60 |
| Output $ / M tokens | $4.68 | $2.40 |
| Results tracked | 41 | 30 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Nemotron 3 Ultra leads
Inkling: 34.5 (#234), Nemotron 3 Ultra: 38.1 (#182)
| Benchmark | Inkling | Nemotron 3 Ultra |
|---|---|---|
| FrontierCode | 14% | 13.6% |
| SciCode | 47% | 40.3% |
| WeirdML | 32.3% | 43.5% |
| LMArena Coding | 1464 | 1468 |
| LMArena WebDev | 1413 | — |
| FrontierSWE | 4.1% | — |
| ALE-Bench | 946 | — |
Agentic & Tool Use Inkling leads
Inkling: 29.6 (#85), Nemotron 3 Ultra: 23.4 (#126)
| Benchmark | Inkling | Nemotron 3 Ultra |
|---|---|---|
| APEX-Agents | 33.8% | 22.7% |
| τ²-bench Banking | 25% | — |
Reasoning Inkling leads
Inkling: 40.4 (#56), Nemotron 3 Ultra: 31.0 (#82)
| Benchmark | Inkling | Nemotron 3 Ultra |
|---|---|---|
| CritPt | 5.4% | 3.1% |
| Chess Puzzles | 21% | 12% |
| LMArena Hard Prompts | 1451 | 1452 |
| DTBench | 87.5% | 90.1% |
| LMCA | 37.6% | 36.9% |
| Epoch Capabilities Index | 148.54 | 146.17 |
| ARC-AGI-2 | 36.5% | — |
| SimpleBench | 50% | — |
| ARC-AGI-1 | 79.5% | — |
| Mystery Game Puzzles | — | 20% |
Math Nemotron 3 Ultra leads
Inkling: 31.3 (#225), Nemotron 3 Ultra: 35.3 (#187)
| Benchmark | Inkling | Nemotron 3 Ultra |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 88.9% | 86.7% |
| ProofBench | 0% | 2% |
| LMArena Math | 1479 | 1457 |
| FrontierMath (Tiers 1-3) | 33.3% | — |
| FrontierMath Tier 4 | 4.9% | — |
Knowledge Inkling leads
Inkling: 55.1 (#49), Nemotron 3 Ultra: 52.5 (#61)
| Benchmark | Inkling | Nemotron 3 Ultra |
|---|---|---|
| GPQA Diamond | 88.3% | 85.4% |
| LMArena Expert | 1465 | 1472 |
| SimpleQA Verified | 40.3% | — |
Multilingual Too close to call
Inkling: 54.0 (#52), Nemotron 3 Ultra: 53.5 (#64)
| Benchmark | Inkling | Nemotron 3 Ultra |
|---|---|---|
| LMArena Non-English | 1434 | 1427 |
| LMArena Chinese | 1490 | 1497 |
| LMArena French | 1458 | 1472 |
| LMArena German | 1446 | 1471 |
| LMArena Korean | 1404 | 1386 |
| LMArena Russian | 1429 | 1417 |
| LMArena Spanish | 1448 | 1454 |
| LMArena Japanese | 1429 | — |
Instruction Following Too close to call
Inkling: 75.1 (#71), Nemotron 3 Ultra: 74.7 (#88)
| Benchmark | Inkling | Nemotron 3 Ultra |
|---|---|---|
| LMArena Instruction Following | 1426 | 1418 |
Long Context Too close to call
Inkling: 43.8 (#86), Nemotron 3 Ultra: 43.9 (#81)
| Benchmark | Inkling | Nemotron 3 Ultra |
|---|---|---|
| LMArena Longer Query | 1434 | 1437 |
Writing & Preference Nemotron 3 Ultra leads
Inkling: 65.2 (#51), Nemotron 3 Ultra: 66.3 (#36)
| Benchmark | Inkling | Nemotron 3 Ultra |
|---|---|---|
| LMArena Text | 1441 | 1445 |
| LMArena Creative Writing | 1387 | 1398 |
| EQ-Bench Creative Writing | 1611 | 1692 |
| LMArena Multi-Turn | 1436 | 1411 |
| EQ-Bench 4 | 1226 | — |
Frequently asked questions
Is Inkling better than Nemotron 3 Ultra?
Inkling is the stronger model overall, scoring 44.1 to 42.5 on the Noometry Index. Nemotron 3 Ultra costs 2.5× less per token, which makes it the better buy when Inkling's lead doesn't matter for your workload.
Which is cheaper, Inkling or Nemotron 3 Ultra?
Nemotron 3 Ultra is cheaper. It lists at $0.60 per million input tokens and $2.40 per million output tokens; Inkling lists at $1.87 and $4.68.
Is Inkling or Nemotron 3 Ultra better for coding?
Nemotron 3 Ultra scores higher on coding benchmarks: 38.1 versus 34.5 in the Noometry coding category.
Which has the bigger context window?
Nemotron 3 Ultra does, with 262K tokens against 66K.
How many benchmarks do Inkling and Nemotron 3 Ultra share?
29 benchmarks have published results for both models. Inkling has 41 scored results on Noometry and Nemotron 3 Ultra has 30.