Model comparison
Inkling vs Nemotron 3 Super
Inkling is the stronger model overall, scoring 44.1 to 39.6 on the Noometry Index. Nemotron 3 Super costs 15× less per token, which makes it the better buy when Inkling's lead doesn't matter for your workload.
Last verified . 19 shared benchmarks.
NVIDIASummary
- They share 19 benchmarks with published results for both. Inkling scores higher in 6 categories and Nemotron 3 Super in 2 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Inkling leads 40.6 to 17.9.
- The biggest single-benchmark swing is SciCode: 47% for Inkling and 36% for Nemotron 3 Super.
- Nemotron 3 Super is cheaper at $0.08 / $0.45 per million input/output tokens, against $1.87 / $4.68 for Inkling.
- Nemotron 3 Super accepts more context: 262K tokens versus 66K.
Side by side
| Inkling | Nemotron 3 Super | |
|---|---|---|
| Provider | Thinking Machines Lab | NVIDIA |
| Noometry Index | 44.1 | 39.6 |
| Released | 2026-07-15 | 2026-03-11 |
| Weights | Open | Open |
| Context window | 66K | 262K |
| Max output | 66K | 131K |
| Input $ / M tokens | $1.87 | $0.08 |
| Output $ / M tokens | $4.68 | $0.45 |
| Results tracked | 41 | 21 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Nemotron 3 Super leads
Inkling: 34.1 (#229), Nemotron 3 Super: 38.9 (#153)
| Benchmark | Inkling | Nemotron 3 Super |
|---|---|---|
| SciCode | 47% | 36% |
| WeirdML | 32.3% | 38% |
| LMArena Coding | 1463 | 1403 |
| ALE-Bench | 946 | 213.9 |
| FrontierCode | 14% | — |
| LMArena WebDev | 1412 | — |
| FrontierSWE | 4.1% | — |
Agentic & Tool Use Not comparable
Inkling: 29.5 (#76), Nemotron 3 Super: —
| Benchmark | Inkling | Nemotron 3 Super |
|---|---|---|
| APEX-Agents | 33.8% | — |
| τ²-bench Banking | 25% | — |
Reasoning Inkling leads
Inkling: 40.6 (#48), Nemotron 3 Super: 17.9 (#274)
| Benchmark | Inkling | Nemotron 3 Super |
|---|---|---|
| CritPt | 5.4% | 3.1% |
| LMArena Hard Prompts | 1452 | 1386 |
| ARC-AGI-2 | 36.5% | — |
| SimpleBench | 50% | — |
| NYT Connections (extended) | — | 15.4% |
| ARC-AGI-1 | 79.5% | — |
| Chess Puzzles | 21% | — |
| DTBench | 87.5% | — |
| LMCA | 37.6% | — |
| Epoch Capabilities Index | 148.6 | — |
Math Nemotron 3 Super leads
Inkling: 31.2 (#198), Nemotron 3 Super: 37.6 (#95)
| Benchmark | Inkling | Nemotron 3 Super |
|---|---|---|
| LMArena Math | 1482 | 1381 |
| FrontierMath (Tiers 1-3) | 33.3% | — |
| FrontierMath Tier 4 | 4.9% | — |
| MathArena Final-Answer Competitions | — | 60.4% |
| OTIS Mock AIME 2024-2025 | 88.9% | — |
| ProofBench | 0% | — |
Knowledge Inkling leads
Inkling: 55.2 (#44), Nemotron 3 Super: 38.2 (#136)
| Benchmark | Inkling | Nemotron 3 Super |
|---|---|---|
| LMArena Expert | 1468 | 1397 |
| GPQA Diamond | 88.3% | — |
| SimpleQA Verified | 40.3% | — |
Multilingual Inkling leads
Inkling: 54.2 (#49), Nemotron 3 Super: 48.6 (#139)
| Benchmark | Inkling | Nemotron 3 Super |
|---|---|---|
| LMArena Non-English | 1434 | 1356 |
| LMArena Chinese | 1495 | 1430 |
| LMArena French | 1453 | 1404 |
| LMArena German | 1456 | 1341 |
| LMArena Russian | 1430 | 1336 |
| LMArena Spanish | 1442 | 1417 |
| LMArena Japanese | 1428 | — |
| LMArena Korean | 1401 | — |
Instruction Following Inkling leads
Inkling: 75.3 (#65), Nemotron 3 Super: 71.4 (#149)
| Benchmark | Inkling | Nemotron 3 Super |
|---|---|---|
| LMArena Instruction Following | 1427 | 1347 |
Long Context Inkling leads
Inkling: 43.9 (#79), Nemotron 3 Super: 41.5 (#134)
| Benchmark | Inkling | Nemotron 3 Super |
|---|---|---|
| LMArena Longer Query | 1436 | 1362 |
Writing & Preference Inkling leads
Inkling: 65.5 (#48), Nemotron 3 Super: 56.1 (#135)
| Benchmark | Inkling | Nemotron 3 Super |
|---|---|---|
| LMArena Text | 1443 | 1378 |
| LMArena Creative Writing | 1390 | 1317 |
| LMArena Multi-Turn | 1436 | 1369 |
| EQ-Bench Creative Writing | 1611 | — |
| EQ-Bench 4 | 1226 | — |
Frequently asked questions
Is Inkling better than Nemotron 3 Super?
Inkling is the stronger model overall, scoring 44.1 to 39.6 on the Noometry Index. Nemotron 3 Super costs 15× 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 Super?
Nemotron 3 Super is cheaper. It lists at $0.08 per million input tokens and $0.45 per million output tokens; Inkling lists at $1.87 and $4.68.
Is Inkling or Nemotron 3 Super better for coding?
Nemotron 3 Super scores higher on coding benchmarks: 38.9 versus 34.1 in the Noometry coding category.
Which has the bigger context window?
Nemotron 3 Super does, with 262K tokens against 66K.
How many benchmarks do Inkling and Nemotron 3 Super share?
19 benchmarks have published results for both models. Inkling has 41 scored results on Noometry and Nemotron 3 Super has 21.