Nemotron 3 Ultra vs Inkling
Inkling leads Nemotron 3 Ultra on coding, 31.5 to 30.0 on the Vibe Coding Index, and Nemotron 3 Ultra costs 1.7x less per output token ($2.40 vs $4.05 per million). Updated October 22, 2026; 67 benchmarks have a result for both.
Nemotron 3 Ultra or Inkling: which to pick
Nemotron 3 Ultra and Inkling are level for coding, 30.0 against 31.5 on the Vibe Coding Index, so price and context decide.
Pick Nemotron 3 Ultra for output tokens 1.7x cheaper ($2.40 vs $4.05 per million) and work measured by LiveCodeBench (Vals AI) or Terminal-Bench 2.1 (Vals AI), where it leads. Pick Inkling for a larger context window (524K against 262K tokens) and work measured by AA Coding Index or SWE-bench Verified (Vals AI), where it leads.
- Shared results
- 67
- Nemotron 3 Ultra only
- 4
- Inkling only
- 39
- Like-for-like categories
- 8 of 9
Only benchmarks that tested both models count toward a pick; a result only one model has is shown but not compared.
Which one for your work
Each card is decided by one number both models carry. Where only one has it, the card says so and makes no call.
Coding
Vibe Coding Index, six public coding benchmarks
No call
Level: 30.0 against 31.5, inside 2 points.
3 and 6 benchmarks behind the scores
Intelligence
AA Intelligence Index
Inkling
Inkling leads, 25.0 against 22.9.
Same board, both measured
Agentic work
AA Agentic Index
Inkling
Inkling leads, 22.5 against 20.1.
Same board, both measured
Reasoning
GPQA Diamond
Inkling
Inkling leads, 87.1% against 86.1%.
Same board, both measured
Knowledge
MMLU Pro
Inkling
Inkling leads, 86.3% against 85.8%.
Same board, both measured
Web design
Design Arena: websites
Inkling
Inkling leads, 1224 against 1145.
Same board, both measured
General
Vals Index
No call
Only Inkling has a Vals Index result, 29.0%.
Directional only
Long documents
Listed context window
Inkling
Inkling takes 524K tokens against 262K.
Listed by OpenRouter
Everyday cost
1K input + 500 output tokens
Nemotron 3 Ultra
Nemotron 3 Ultra $0.0018, Inkling $0.0030 per turn.
List prices
Cost per solved task
One SWE-bench Verified task, solved
No call
Only Inkling has the number, $0.76.
Directional only
At a glance
- Coding index
- Nemotron 3 Ultra 30.0 from 3 benchmarks, Inkling 31.5 from 6 benchmarks.
- Widest category
- Agentic work, 9.2 points apart on average over 6 shared benchmarks.
- Widest single gap
- Legal Research Bench (Vals AI): Nemotron 3 Ultra 15.4% against Inkling 28.4%, 13.0 points apart on a 0 to 100 scale.
- Output price
- Nemotron 3 Ultra $2.40 against Inkling $4.05 per million output tokens, 1.7x apart.
- Context window
- Nemotron 3 Ultra takes 262K tokens, Inkling 524K.
Benchmark by benchmark
First the score: boards each model leads, per category, where both have a result. Then every board, a category to a fold; bars are each result on a 0 to 100 scale, so compare the two in a row, not one row with another.
| Category | Nemotron 3 Ultra | Bars | Inkling | Ties |
|---|---|---|---|---|
| Coding | 5 | 6 | 1 tie | |
| Intelligence | 0 | 2 | ||
| Agentic work | 0 | 6 | ||
| Reasoning | 3 | 4 | ||
| Math | 2 | 2 | ||
| Knowledge | 3 | 6 | ||
| Web design | 1 | 7 | ||
| General | 12 | 7 | ||
| All shared boards | 26 | 40 | 1 tie |
CodingLike-for-likeSix of these boards feed the Vibe Coding Index56
| Benchmark | Nemotron 3 Ultra | Inkling | Leader |
|---|---|---|---|
| LiveBench coding | Not measured | 64.0% | Only Inkling measured |
| LMArena WebDev | Not measured | 1414 | Only Inkling measured |
| AA Coding Index | 49.3 | 52.1 | Inkling |
| LiveCodeBench (Vals AI) | 86.0% | 85.5% | Nemotron 3 Ultra |
| SWE-bench Verified (Vals AI) | 69.0% | 77.6% | Inkling |
| Terminal-Bench 4 (Vals AI) | Not measured | 0.5% | Only Inkling measured |
| Vibe Code Bench (Vals AI) | 7.6% | 19.2% | Inkling |
| Code migration (Vals AI) | 4.9% | 11.8% | Inkling |
| ProgramBench (Vals AI) | 0.0% | 0.0% | Tie |
| IOI (Vals AI) | Not measured | 14.9% | Only Inkling measured |
| Terminal-Bench 2.1 (Vals AI) | 50.9% | 47.6% | Nemotron 3 Ultra |
| Vibe Code Bench 1-100 (Vals AI) | Not measured | 7.3% | Only Inkling measured |
| SkillsBench (Vals AI) | Not measured | 26.1% | Only Inkling measured |
| CyberBench (Vals AI) | Not measured | 66.8% | Only Inkling measured |
| SciCode | 40.3% | 47.0% | Inkling |
| Terminal-Bench Hard (Artificial Analysis run) | 36.4% | Not measured | Only Nemotron 3 Ultra measured |
| FrontierCode | 13.6% | 14.0% | Inkling |
| FrontierSWE | Not measured | 4.1% | Only Inkling measured |
| WeirdML | 43.5% | 32.3% | Nemotron 3 Ultra |
| ALE-Bench | Not measured | 946.0 | Only Inkling measured |
| LMArena Text: coding | 1468 | 1466 | Nemotron 3 Ultra |
| LMArena Text: software and IT | 1468 | 1465 | Nemotron 3 Ultra |
| LMArena WebDev: React | Not measured | 1404 | Only Inkling measured |
| LMArena WebDev: HTML | Not measured | 1399 | Only Inkling measured |
| SWE-Bench Pro V2 | Not measured | 89.9% | Only Inkling measured |
| Neither model has a result on the other 18 boards here | |||
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index02
| Benchmark | Nemotron 3 Ultra | Inkling | Leader |
|---|---|---|---|
| AA Intelligence Index | 22.9 | 25.0 | Inkling |
| Epoch Capabilities Index | 146.0 | 148.4 | Inkling |
Agentic workLike-for-likePublic boards, shown as published, kept out of the index06
| Benchmark | Nemotron 3 Ultra | Inkling | Leader |
|---|---|---|---|
| AA Agentic Index | 20.1 | 22.5 | Inkling |
| Design Arena: agents building web apps | Not measured | 1196 | Only Inkling measured |
| LiveBench research | Not measured | 59.9% | Only Inkling measured |
| LiveBench browser | Not measured | 16.0% | Only Inkling measured |
| LiveBench long agentic | Not measured | 14.1% | Only Inkling measured |
| SREBench (Vals AI) | Not measured | 0.0% | Only Inkling measured |
| Finance Agent (Vals AI) | 37.7% | 46.6% | Inkling |
| Tax Agent Bench (Vals AI) | Not measured | 15.3% | Only Inkling measured |
| Legal Research Bench (Vals AI) | 15.4% | 28.4% | Inkling |
| Harvey's Legal Agent Benchmark (Vals AI) | Not measured | 3.8% | Only Inkling measured |
| Excel Modeling Benchmark (Vals AI) | 31.7% | 38.4% | Inkling |
| τ²-Bench Telecom | 83.3% | Not measured | Only Nemotron 3 Ultra measured |
| APEX-Agents | 22.7% | 33.8% | Inkling |
| MCP Atlas | 63.1% | 76.0% | Inkling |
| Neither model has a result on the other 28 boards here | |||
ReasoningLike-for-likePublic boards, shown as published, kept out of the index34
| Benchmark | Nemotron 3 Ultra | Inkling | Leader |
|---|---|---|---|
| GPQA Diamond | 86.1% | 87.1% | Inkling |
| LiveBench reasoning | Not measured | 78.3% | Only Inkling measured |
| MMMU Pro (Vals AI) | Not measured | 78.5% | Only Inkling measured |
| GPQA Diamond (Artificial Analysis run) | 86.7% | 87.2% | Inkling |
| Humanity's Last Exam (Artificial Analysis run) | 28.4% | 31.9% | Inkling |
| GPQA Diamond (Epoch AI run) | 85.4% | 88.3% | Inkling |
| ARC-AGI | Not measured | 79.5% | Only Inkling measured |
| ARC-AGI-2 | Not measured | 36.5% | Only Inkling measured |
| SimpleBench | Not measured | 50.0% | Only Inkling measured |
| Chess puzzles | Not measured | 21.0% | Only Inkling measured |
| Mystery game puzzles | 20.0% | Not measured | Only Nemotron 3 Ultra measured |
| LMArena Text: hard prompts | 1452 | 1451 | Nemotron 3 Ultra |
| LMArena Text: expert | 1472 | 1466 | Nemotron 3 Ultra |
| LMArena Text: hard prompts in English | 1453 | 1448 | Nemotron 3 Ultra |
| Humanity's Last Exam (Scale run) | Not measured | 11.2% | Only Inkling measured |
| Neither model has a result on the other 8 boards here | |||
MathLike-for-likePublic boards, shown as published, kept out of the index22
| Benchmark | Nemotron 3 Ultra | Inkling | Leader |
|---|---|---|---|
| OTIS Mock AIME | 86.7% | 88.9% | Inkling |
| FrontierMath, tiers 1 to 3 | Not measured | 33.3% | Only Inkling measured |
| FrontierMath, tier 4 | Not measured | 4.9% | Only Inkling measured |
| ProofBench | 2.0% | 0.0% | Nemotron 3 Ultra |
| LMArena Text: math | 1457 | 1479 | Inkling |
| LMArena Text: mathematical work | 1461 | 1460 | Nemotron 3 Ultra |
| Neither model has a result on the other 8 boards here | |||
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index36
| Benchmark | Nemotron 3 Ultra | Inkling | Leader |
|---|---|---|---|
| MMLU Pro | 85.8% | 86.3% | Inkling |
| LegalBench (Vals AI) | 82.1% | 83.0% | Inkling |
| CorpFin (Vals AI) | 65.5% | 68.6% | Inkling |
| TaxEval (Vals AI) | 73.1% | 75.3% | Inkling |
| MortgageTax (Vals AI) | Not measured | 64.0% | Only Inkling measured |
| MedCode (Vals AI) | 38.6% | 41.2% | Inkling |
| SimpleQA Verified | Not measured | 40.3% | Only Inkling measured |
| LMArena Text: science | 1465 | 1455 | Nemotron 3 Ultra |
| LMArena Text: medicine and healthcare | 1456 | 1447 | Nemotron 3 Ultra |
| LMArena Text: legal and government | 1452 | 1448 | Nemotron 3 Ultra |
| LMArena Text: business and finance | 1434 | 1438 | Inkling |
| Neither model has a result on the other 9 boards here | |||
Web designLike-for-likePublic boards, shown as published, kept out of the index17
| Benchmark | Nemotron 3 Ultra | Inkling | Leader |
|---|---|---|---|
| Design Arena: websites | 1145 | 1224 | Inkling |
| Design Arena: UI components | 1135 | 1199 | Inkling |
| Design Arena: game dev | 1138 | 1165 | Inkling |
| Design Arena: data viz | 1141 | 1157 | Inkling |
| Design Arena: 3D | 1131 | 1160 | Inkling |
| Design Arena: all code categories | 1146 | 1205 | Inkling |
| Design Arena: SVG | 1069 | 1108 | Inkling |
| Design Arena: ASCII art | 1100 | 1089 | Nemotron 3 Ultra |
| LiveBench design | Not measured | 58.3% | Only Inkling measured |
GeneralLike-for-likePublic boards, shown as published, kept out of the index127
| Benchmark | Nemotron 3 Ultra | Inkling | Leader |
|---|---|---|---|
| Vals Index | Not measured | 29.0% | Only Inkling measured |
| LiveBench data analysis | Not measured | 72.8% | Only Inkling measured |
| Public Benefits Bench (Vals AI) | Not measured | 58.5% | Only Inkling measured |
| SAGE (Vals AI) | Not measured | 36.5% | Only Inkling measured |
| MedScribe (Vals AI) | Not measured | 85.4% | Only Inkling measured |
| Vals Multimodal Index | Not measured | 49.4% | Only Inkling measured |
| IFBench | 81.4% | Not measured | Only Nemotron 3 Ultra measured |
| Long context reasoning (AA-LCR) | 79.3% | 77.3% | Nemotron 3 Ultra |
| DTBench | 90.1% | 87.5% | Nemotron 3 Ultra |
| LMCA | 36.9% | 37.6% | Inkling |
| LMArena Text | 1445 | 1442 | Nemotron 3 Ultra |
| LMArena Text: instruction following | 1418 | 1425 | Inkling |
| LMArena Text: creative writing | 1399 | 1390 | Nemotron 3 Ultra |
| LMArena Text: English | 1455 | 1442 | Nemotron 3 Ultra |
| LMArena Text: non-English | 1427 | 1434 | Inkling |
| LMArena Text: Chinese | 1497 | 1493 | Nemotron 3 Ultra |
| LMArena Text: German | 1472 | 1445 | Nemotron 3 Ultra |
| LMArena Text: Japanese | Not measured | 1429 | Only Inkling measured |
| LMArena Text: Korean | 1384 | 1403 | Inkling |
| LMArena Text: French | 1472 | 1455 | Nemotron 3 Ultra |
| LMArena Text: Spanish | 1455 | 1449 | Nemotron 3 Ultra |
| LMArena Text: Russian | 1417 | 1430 | Inkling |
| LMArena Text: Polish | 1385 | 1450 | Inkling |
| LMArena Text: multi-turn | 1411 | 1437 | Inkling |
| LMArena Text: longer queries | 1437 | 1435 | Nemotron 3 Ultra |
| LMArena Text: writing and language | 1406 | 1404 | Nemotron 3 Ultra |
| LMArena Text: entertainment and media | 1397 | 1394 | Nemotron 3 Ultra |
| EQ-Bench 4 | Not measured | 1226 | Only Inkling measured |
| Neither model has a result on the other 20 boards here | |||
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
| Benchmark | Nemotron 3 Ultra | Inkling | Leader |
|---|---|---|---|
| Neither model has a result on 5 boards here | |||
What each workload costs
Three fixed token mixes at each model's listed price, so per-token rates become a bill.
Chat turn
1K input + 500 output tokens
Nemotron 3 Ultra costs less
Repository review
50K input + 3K output tokens
Nemotron 3 Ultra costs less
Agent loop
200K input + 10K output tokens
Nemotron 3 Ultra costs less
Listed OpenRouter rates, without cache or batch discounts. A workload that does not fit a model's context window would need splitting.
Specification differences
The leading value in a row is marked; a row with a missing value names no leader. Our two rows are marked in red.
| Specification | NVIDIA Nemotron 3 Ultra | Thinking Machines Inkling |
|---|---|---|
| Maker | NVIDIA | Thinking Machines |
| Weights | Open weights | Open weights |
| Released | June 4, 2026 | July 17, 2026 |
| Context window | 262K | 524K |
| Input $/Mtok | $0.60 | $1.00 |
| Output $/Mtok | $2.40 | $4.05 |
| Coding index | 30.0 | 31.5 |
| Independent measurements behind it | 3 | 6 |
| Cost per solved task | Not computable | $0.76 |
| Source | OpenRouter listing | OpenRouter listing |
Questions
Is Nemotron 3 Ultra better than Inkling for coding?
Nemotron 3 Ultra and Inkling are level for coding, 30.0 against 31.5 on the Vibe Coding Index, so price and context decide. Pick Nemotron 3 Ultra for output tokens 1.7x cheaper ($2.40 vs $4.05 per million) and work measured by LiveCodeBench (Vals AI) or Terminal-Bench 2.1 (Vals AI), where it leads. Pick Inkling for a larger context window (524K against 262K tokens) and work measured by AA Coding Index or SWE-bench Verified (Vals AI), where it leads.
Which is cheaper, Nemotron 3 Ultra or Inkling?
Nemotron 3 Ultra costs $0.60 per million input tokens and $2.40 per million output tokens; Inkling costs $1.00 and $4.05. Cost per solved task is known for Inkling only.
Which has the bigger context window, Nemotron 3 Ultra or Inkling?
Inkling, with 524K tokens against 262K.
How many benchmarks is the Nemotron 3 Ultra vs Inkling comparison based on?
67 benchmarks have a result for both models as of October 22, 2026. Nemotron 3 Ultra's coding index rests on 3 independent benchmarks and Inkling's on 6. Where only one model has a result, the page says so instead of naming a winner.