Inkling vs Nemotron 3.5 Super VL 120B A12B
Full profiles: Inkling (Thinking Machines) and Nemotron 3.5 Super VL 120B A12B (NVIDIA): every benchmark, price and rank.
Updated · Data from 7 sources
As of 9 Nov 2026, Inkling and Nemotron 3.5 Super VL 120B A12B are neck and neck on quality (50.1 vs 50.3 on the BenchLeader Index, inside the ±10.6 margin), and Nemotron 3.5 Super VL 120B A12B is 3.3× cheaper at $0.788 vs $2.57 per 1M tokens blended. Choose Inkling for agents & tools (55 vs 24) and multimodal (56 vs 54); choose Nemotron 3.5 Super VL 120B A12B for maths (59 vs 45), reasoning (59 vs 47), and coding (48 vs 38).
Key facts (as of 9 Nov 2026)
- Inkling scores 50.1 on the BenchLeader Index and Nemotron 3.5 Super VL 120B A12B scores 50.3 (BenchLeader, 9 Nov 2026).
- Inkling costs $2.57 per 1M tokens blended (list price, 9 Nov 2026).
- Nemotron 3.5 Super VL 120B A12B costs $0.788 per 1M tokens blended (list price, 9 Nov 2026).
Inkling benchmarks xhigh Thinking Machines
Nemotron 3.5 Super VL 120B A12B benchmarks NVIDIAAdd a third modelWhich to pick for…
The pick's numbers come first. Category scores on the BenchLeader scale (50 is the average model, 15 points one standard deviation). High confidence means a lead of half a standard deviation or more; under one point is called either.
| Use case | Pick | Why | Confidence |
|---|---|---|---|
| Coding | Nemotron 3.5 Super VL 120B A12B | 48 vs 38 category score | High |
| Agents & tool use | Inkling | 55 vs 24 category score | High |
| Maths | Nemotron 3.5 Super VL 120B A12B | 59 vs 45 category score | High |
| Reasoning | Nemotron 3.5 Super VL 120B A12B | 59 vs 47 category score | High |
| Images and other media | Inkling | 56 vs 54 category score | Medium |
| Chat quality (human preference) | – | Only Nemotron 3.5 Super VL 120B A12B has an Arena rating (59 category score) | Not comparable |
| Lowest cost | Nemotron 3.5 Super VL 120B A12B | $0.788 vs $1.76 per 1M blended; agentic coding session $0.040 vs $0.076 | – |
| High volume on a budget | Neither flagship | A smaller model is usually the better fit: Cheapest good LLMs, Fastest LLMs | – |
Key differences
- Inkling (xhigh) and Nemotron 3.5 Super VL 120B A12B are neck and neck on quality (50.1 vs 50.3, within the ±10.6 margin).
- Inkling (xhigh) is stronger in agents & tools, knowledge, long context, multimodal.
- Nemotron 3.5 Super VL 120B A12B is stronger in coding, maths, reasoning, human preference.
- Nemotron 3.5 Super VL 120B A12B is 3.3× cheaper ($0.788 vs $2.57 per 1M blended).
At a glance
| Spec | Inkling | Nemotron 3.5 Super VL 120B A12B |
|---|---|---|
| Developer | Thinking Machines | NVIDIA |
| Released | 15 Jul 2026 | 15 Oct 2026 |
| Context window (tokens) | 65,536 | 1,048,576 |
| Compared at | Inkling (xhigh) | Nemotron 3.5 Super VL 120B A12B |
| Open weights | Yes | Yes |
| Price in / out / cached input per 1M | $1.00 / $4.05 / $0.374 | $0.570 / $1.44 / $0.114 |
Category by category
Category scores on the BenchLeader scale: 50 is the average evaluated model, 15 points is one standard deviation. A gap under one point is called level.
| Category | Inkling (xhigh) | Nemotron 3.5 Super VL 120B A12B | Leader |
|---|---|---|---|
| Reasoning | 47.1 | 58.9 | Nemotron 3.5 Super VL 120B A12B |
| Coding | 37.8 | 48.1 | Nemotron 3.5 Super VL 120B A12B |
| Agents & tools | 55.4 | 23.9 | Inkling |
| Maths | 45.0 | 59.2 | Nemotron 3.5 Super VL 120B A12B |
| Knowledge | 46.5 | – | not comparable |
| Human preference | – | 59.1 | not comparable |
| Multimodal | 55.8 | 53.5 | Inkling |
| Long context | 63.3 | – | not comparable |
On one chart
Each domain on the index's own scale, where 50 is an average model. Click a domain to see the benchmarks behind it. Pick up to 5 models from the list.
Not drawn, measured for fewer than two of these models: Knowledge, Instruction following, Long context, Human preference.
Where sources disagree
Across the 0 benchmarks both have, Inkling leads 0, Nemotron 3.5 Super VL 120B A12B leads 0. Independent sources split: Inkling leads on Arena Text; Nemotron 3.5 Super VL 120B A12B leads on Vals Index.
| Overall index | Inkling (xhigh) | Nemotron 3.5 Super VL 120B A12B | Leader |
|---|---|---|---|
| BenchLeader Index | 50.1 | 50.3 | neck and neck |
| Vals Index | 29.0† | 29.8 | Nemotron 3.5 Super VL 120B A12B |
| Arena Text | 1443† | 1415 | Inkling |
Each publisher runs its own tasks, harnesses and effort settings, so their overall verdicts can differ; BenchLeader reconciles them on one scale. † measured at another effort setting of the same model (the publisher did not test the one compared here).
Price, speed and latency
List prices blended at three input tokens per output token. Per-task prices use the token counts on the methodology page. Cost, tokens and time per benchmark task are measured: Artificial Analysis’ averages over its whole Intelligence Index run.
| Measure | Inkling (xhigh) | Nemotron 3.5 Super VL 120B A12B |
|---|---|---|
| Blended price per 1M tokens | $1.76 | $0.788 |
| Input / output per 1M | $1.00 / $4.05 | $0.570 / $1.44 |
| Cached input per 1M | $0.374 | $0.114 |
| Price per task: chat reply | $0.0016 | $0.0007 |
| Price per task: summarize a 30-page report | $0.014 | $0.0077 |
| Price per task: code edit | $0.012 | $0.0056 |
| Price per task: agentic coding session | $0.076 | $0.040 |
| Price per task: structured extraction | $0.0028 | $0.0014 |
| Agentic coding session, input mostly cached | $0.048 | $0.019 |
| Cost per benchmark task (AA Index) | $1.35 | – |
| Output tokens per task (AA Index) | 51k | – |
| Of which reasoning | 34k | – |
| Time per task (AA Index) | 8 min | – |
| Output speed | 107 tok/s | – |
| Time to first answer | 17.7 s | – |
| Time to a complete answer | 46.5 s | – |
Each model is compared at its best-ranked reasoning-effort setting. Inkling (xhigh) scores 50.1 and takes 18 s to its first answer token; Inkling scores 47.7 in 3 s. Nemotron 3.5 Super VL 120B A12B scores 50.3. Higher effort means more thinking before the answer: better scores, longer waits and more tokens billed. How reasoning effort changes scores and cost.
Every benchmark
122 benchmarks, grouped by category; each name links to the board and its leaders. † measured at another effort setting of the same model, shown for reference and not counted as a win.
| Metric | Publisher | Inkling (xhigh) | Nemotron 3.5 Super VL 120B A12B |
|---|---|---|---|
| BenchLeader Index | BenchLeader | 50.1 | 50.3 |
| Agents & tools score | BenchLeader | 55.4 | 23.9 |
| Coding score | BenchLeader | 37.8 | 48.1 |
| Knowledge score | BenchLeader | 46.5 | – |
| Long context score | BenchLeader | 63.3 | – |
| Maths score | BenchLeader | 45.0 | 59.2 |
| Multimodal score | BenchLeader | 55.8 | 53.5 |
| Reasoning score | BenchLeader | 47.1 | 58.9 |
| Human preference score | BenchLeader | – | 59.1 |
| Blended price $/M | BenchLeader | $1.76 | $0.788 |
| Output speed | BenchLeader | 107 tok/s | – |
| Time to first answer | BenchLeader | 17.7 s | – |
| Context window | BenchLeader | 65,536 tokens | 1,048,576 tokens |
| Reasoning | |||
| GPQA Diamond | Epoch AI Benchmarking Hub | 88.3% | – |
| SimpleBench | SimpleBench | 50.0%† | – |
| Arena Hard Prompts | Arena (LMArena) | 1463† | 1432 |
| LiveBench 2.0 Reasoning | LiveBench | 78.3%† | – |
| LiveBench 1.0 Reasoning | LiveBench | 78.3% | – |
| GPQA Diamond (AA) | Artificial Analysis | 87.2% | – |
| Humanity's Last Exam (AA) | Artificial Analysis | 31.9% | – |
| HLE-Diamond (AA) | Artificial Analysis | 10.6% | – |
| GPQA Diamond (Vals) | Vals AI | 87.1%† | – |
| ARC-AGI-1 | ARC Prize | 79.5%† | – |
| ARC-AGI-2 | ARC Prize | 36.5%† | – |
| CritPt | Artificial Analysis | 5.4% | – |
| CritPt v2 hard subset (AA) | Artificial Analysis | 18.0% | – |
| Chess Puzzles | Epoch AI Benchmarking Hub | 21.0% | – |
| LMCA | Epoch AI Benchmarking Hub | 37.6% | – |
| DTBench | Epoch AI Benchmarking Hub | 87.5% | – |
| Coding | |||
| SciCode | SciCode | 47.0% | – |
| WeirdML | WeirdML | 32.3%† | – |
| FrontierCode | Cognition | 14.0%† | – |
| Arena Coding | Arena (LMArena) | 1494† | 1460 |
| Arena WebDev | Arena (LMArena) | 1414† | 1399 |
| LiveBench 2.0 Coding | LiveBench | 48.7%† | – |
| LiveBench 1.0 Coding | LiveBench | 71.0% | – |
| LiveCodeBench | Vals AI | 85.5%† | – |
| IOI | Vals AI | 14.9%† | – |
Show all 122 benchmarks
| Metric | Publisher | Inkling (xhigh) | Nemotron 3.5 Super VL 120B A12B |
|---|---|---|---|
| Coding | |||
| SWE-bench (Vals) | Vals AI | 77.6%† | – |
| Code Migration | Vals AI | 11.8%† | – |
| Code Migration v1.1 | Vals AI | 8.8%† | 16.4%† |
| ProgramBench | Vals AI | 0.0%† | – |
| Vibe Code Bench 1-100 | Vals AI | 7.3%† | – |
| Vibe Code Bench v1.1 | Vals AI | 19.2%† | 3.7%† |
| ALE-Bench | Epoch AI Benchmarking Hub | 946† | – |
| FrontierSWE | Epoch AI Benchmarking Hub | 4.1% | – |
| Terminal-Bench 4.0 (AA) | Artificial Analysis | 1.0% | – |
| Terminal-Bench 2.1 (AA) | Artificial Analysis | 55.1% | – |
| GameBench 2 Overall | DuelLab GameBench 2 | 25.4† | – |
| GameBench 2 by reasoning setting | DuelLab GameBench 2 | 26.6† | – |
| Design Arena Website | Design Arena | 1221† | – |
| FrontierCode 1.1 (AA) | Artificial Analysis | 15.8% | – |
| FrontierCode 2.0 (AA) | Artificial Analysis | 3.9% | – |
| Agents & tools | |||
| APEX-Agents | Mercor | 33.8%† | – |
| Arena Agent | Arena (LMArena) | -17.1† | -16.3 |
| LiveBench Long Agentic | LiveBench | 9.8%† | – |
| LiveBench Browser | LiveBench | 12.0%† | – |
| LiveBench Agentic Coding | LiveBench | 49.4% | – |
| Terminal-Bench 2.1 (Vals) | Vals AI | 47.6%† | – |
| MCP Atlas | Scale AI SEAL | 76.0% | – |
| τ³-bench Banking | τ²-bench | 25.0%† | – |
| GDPval-AA v2.1 | Artificial Analysis | 1080 | – |
| τ³-Banking (AA) | Artificial Analysis | 29.1% | – |
| Analyst Agent (AA) | Artificial Analysis | 23.8% | – |
| CyberBench | Vals AI | 66.8%† | – |
| Finance Agent v2 | Vals AI | 46.6%† | – |
| Finance Agent v2.1 | Vals AI | 49.9%† | 47.7%† |
| Harvey's Legal Agent Benchmark v1.1 | Vals AI | 3.8%† | 3.8%† |
| Harvey's Legal Agent Benchmark v1 | Vals AI | 2.1%† | – |
| Legal Research Bench | Vals AI | 28.4%† | 10.1%† |
| SkillsBench | Vals AI | 26.1%† | – |
| SREBench | Vals AI | 0.0%† | – |
| Tax Agent Bench | Vals AI | 15.3%† | 11.6%† |
| Terminal-Bench 4.0 (Vals) | Vals AI | 0.5%† | 0.0%† |
| AutomationBench | Artificial Analysis | 5.0% | – |
| GDP.pdf (AA) | Artificial Analysis | 12.8% | – |
| Harvey LAB-AA v1.1 | Harvey | 0.0% | – |
| EnterpriseOps-Gym | Artificial Analysis | 38.0% | – |
| AA-Briefcase v1.1 | Artificial Analysis | 833 | – |
| GDPx | GDPx by SemiAnalysis | 2.3%† | – |
| Design Arena Agentic | Design Arena | 1168† | – |
| Terminal-Bench-Science 0.1 (AA) | Artificial Analysis | 0.0% | – |
| Maths | |||
| FrontierMath Tiers 1–3 | Epoch AI Benchmarking Hub | 33.3% | – |
| FrontierMath Tier 4 | Epoch AI Benchmarking Hub | 4.9% | – |
| OTIS Mock AIME | Epoch AI Benchmarking Hub | 88.9% | – |
| ProofBench | Vals AI | 0.0%† | – |
| LiveBench Mathematics | LiveBench | 88.4% | – |
| Arena Maths | Arena (LMArena) | 1479† | 1417 |
| Knowledge | |||
| SimpleQA Verified | Epoch AI Benchmarking Hub | 40.3% | – |
| LiveBench 2.0 Data Analysis | LiveBench | 72.8%† | – |
| LiveBench Research | LiveBench | 60.3%† | – |
| LiveBench 1.0 Data Analysis | LiveBench | 72.8% | – |
| AA-Omniscience v1.1 | Artificial Analysis | -1.5 | – |
| MMLU-Pro | Vals AI | 86.3%† | – |
| LegalBench | Vals AI | 83.0%† | – |
| CorpFin | Vals AI | 68.6%† | – |
| TaxEval | Vals AI | 75.3%† | – |
| Excel Modeling Benchmark | Vals AI | 38.4%† | – |
| Excel Modeling Benchmark v1.1 | Vals AI | 41.0%† | 52.8%† |
| MedCode | Vals AI | 41.2%† | – |
| MedScribe | Vals AI | 85.4%† | – |
| Public Benefits Bench v1.1 | Vals AI | 58.5%† | – |
| AA-Omniscience v1.1: accuracy | Artificial Analysis | 39.3% | – |
| AA-Omniscience v1.1: non-hallucination | Artificial Analysis | 32.8% | – |
| Instruction following | |||
| LiveBench Language | LiveBench | 73.5% | – |
| LiveBench Instruction Following | LiveBench | 70.1% | – |
| Arena Instruction Following | Arena (LMArena) | 1428† | 1411 |
| Human preference | |||
| Arena Text | Arena (LMArena) | 1443† | 1415 |
| EQ-Bench 4 | EQ-Bench | 1212† | – |
| Arena Creative Writing | Arena (LMArena) | 1389† | 1388 |
| Arena Multi-turn | Arena (LMArena) | 1445† | 1425 |
| Arena Longer Queries | Arena (LMArena) | 1442† | 1412 |
| Multimodal | |||
| Arena Vision | Arena (LMArena) | – | 1214 |
| LiveBench Design | LiveBench | 55.2%† | – |
| MMMU-Pro | Artificial Analysis | 73.5% | – |
| MMMU-Pro (Vals) | Vals AI | 78.5%† | – |
| MortgageTax | Vals AI | 64.0%† | – |
| AudioMC | Scale AI SEAL | 56.6%† | – |
| AudioMC (text output) | Scale AI SEAL | 56.6%† | – |
| Code2Video-AA | Artificial Analysis | 467 | – |
| AA-3D | Artificial Analysis | 591 | – |
| Long context | |||
| AA-LCR v1.1 | Artificial Analysis | 77.3% | – |
| MLCR | Artificial Analysis | 6.1% | – |
| Composite | |||
| Epoch Capabilities Index | Epoch AI Benchmarking Hub | 148.4† | – |
| LiveBench 2.0 | LiveBench | 44.3%† | – |
| LiveBench 1.0 | LiveBench | 71.9% | – |
| AA Intelligence Index v5.0 | Artificial Analysis | 12.5 | – |
| AA Intelligence Index v4.3.2 | Artificial Analysis | 25.0 | – |
| AA Progress Index v1 | Artificial Analysis | 281.4 | – |
| Vals Index | Vals AI | 29.0† | 29.8† |
| Vals Finance Index | Vals AI | 38.3%† | 41.0%† |
| Vals Multimodal Index | Vals AI | 49.4%† | – |
| Vals Router Index | Vals AI | 29.0%† | – |
| Safety & honesty | |||
| FORTRESS | Scale AI SEAL | 25.0% | – |
| Metadata | |||
| AA Openness Index | Artificial Analysis | 38.9 | – |
Data as of 2026-11-09. Best configuration of each model; every score links to its source on the model pages.
DuelLab GameBench 2, release gb2-20261103-eaea31df94e6 (https://benchmarks.duellab.org/releases/gb2-20261103-eaea31df94e6/), CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). Scores mapped to BenchLeader model names; not endorsed by DuelLab.
What changed
- Inkling (xhigh): 1 score revision.
- Nemotron 3.5 Super VL 120B A12B: 1 score revision.
- Nemotron 3.5 Super VL 120B A12B: 1 score revision.
- Nemotron 3.5 Super VL 120B A12B: 1 score revision.
- Inkling (xhigh): BenchLeader Index 49.8 → 50.3.
- Nemotron 3.5 Super VL 120B A12B: 1 score revision.
- Inkling (xhigh): BenchLeader Index 54.4 → 49.8.
- Inkling (xhigh): 5 new results (AA Intelligence Index v5.0, AA Progress Index v1, HLE-Diamond (AA)…); 2 score revisions; cost per task moved from $0.60/M to $1.35/M (127% up).
- Nemotron 3.5 Super VL 120B A12B: BenchLeader Index 53.6 → 50.4.
- Nemotron 3.5 Super VL 120B A12B: 1 new result (Arena Agent).
Related comparisons
Rankings: Best LLM overall · Best LLM for agents · Best cheap LLMs
Inkling vs Nemotron 3.5 Super VL 120B A12B: questions
- Is Inkling better than Nemotron 3.5 Super VL 120B A12B?
- Inkling (xhigh) and Nemotron 3.5 Super VL 120B A12B are neck and neck on quality (50.1 vs 50.3, within the ±10.6 margin). The BenchLeader Index combines every independent quality benchmark; the two are within the margin of error as of 9 Nov 2026, so check the category scores for your use.
- Inkling vs Nemotron 3.5 Super VL 120B A12B: which is cheaper, and will that change?
- Nemotron 3.5 Super VL 120B A12B is cheaper today: $0.788 against $2.57 per million tokens, blended at three input tokens per output token. No price change has been announced for either.
- Is Inkling or Nemotron 3.5 Super VL 120B A12B better for coding?
- Nemotron 3.5 Super VL 120B A12B scores higher in coding (48 vs 38 on the category index, where 50 is average).
- Is Inkling or Nemotron 3.5 Super VL 120B A12B better for agents and tool use?
- Inkling scores higher in agentic tasks (55 vs 24 on the category index, where 50 is average).
- Which is better for long documents?
- Nemotron 3.5 Super VL 120B A12B accepts more context: 1,048,576 against 65,536 tokens for Inkling.
More questions (1)
- When was Nemotron 3.5 Super VL 120B A12B released?
- Nemotron 3.5 Super VL 120B A12B was released on 15 Oct 2026.