Skip to content
VECTOR WIREAI INTELLIGENCE
UTC

Inkling 2 Mini vs Nemotron 3.5 Super VL

Thinking Machines · released

Wins 4 of 4 areas

Coding · Agents · Reasoning · Long documents

NVIDIA · released

Wins 0 of 4 areas

—

Inkling 2 Mini is the stronger all-rounder.

Scores updated · 33 tests both models report · How we compare

Where each one wins

Tests won in each of the four areas where both have results. Each piece is one test, so a longer bar means more evidence; grey means the two scored within a point of each other.

Long documents rests on a single test.

What it costs

Prices per million tokens, roughly 750,000 words. The bars show the cost of a million tokens read plus a million written.

Inkling 2 Mini$0.15 to read · $0.50 to write$0.65Price from Artificial Analysis
Nemotron 3.5 Super VL$0.15 to read · $0.50 to write$0.65Price from Artificial Analysis

The biggest differences

The tests each model wins by the widest margin, up to three each. Scores are out of 100.

Where Inkling 2 Mini pulls ahead

  • Abstract visual puzzles that people can solveARC-AGI-2+47.8points aheadInkling 2 Mini49.3Nemotron 3.5 Super VL1.5
  • Research-level physics problemsCritPt+20.8points aheadInkling 2 Mini31.4Nemotron 3.5 Super VL10.6
  • Multi-step tasks using many real software toolsMCP Atlas+15points aheadInkling 2 Mini82.7Nemotron 3.5 Super VL67.7

Where Nemotron 3.5 Super VL pulls ahead

No clear win on a test scored out of 100.

Every test, side by side

All 33 tests both models report. The winning score is in its model's colour; marks a score checked independently.

CodingInkling 2 Mini

Full coding ranking

AgentsInkling 2 Mini

Full agents ranking

ReasoningInkling 2 Mini
  • ARC-AGI-2Inkling 2 Mini by 47.849.31.5+47.8
  • CritPtInkling 2 Mini by 20.831.410.6+20.8

Full reasoning ranking

Long documentsInkling 2 Mini

Full long documents ranking

Other results25 tests, not counted

Tests outside the eight areas. They are not counted above: several are summary scores built from other tests, or the same test under another name.

  • ARC-AGI-1Inkling 2 Mini by 61.884.522.7+61.8
  • Misbehavior (Internal) ↓Nemotron 3.5 Super VL by 439.852.8+43
  • MRCR v2 8-needle 256K–512KInkling 2 Mini by 40.746.45.7+40.7
  • Covert misbehavior (Internal) ↓Nemotron 3.5 Super VL by 35.18.343.4+35.1
  • SWEBench Pro v2 (Hard)Inkling 2 Mini by 33.877.944.1+33.8
  • ScreenSpot Pro originalInkling 2 Mini by 3376.543.5+33
  • AutomationBench-AA v1.0.6Inkling 2 Mini by 2741.814.8+27
  • DeepSWE 1.1Inkling 2 Mini by 23.548.224.7+23.5
  • BVBInkling 2 Mini by 16.368.452.1+16.3
  • GDPval-AA 2.1Inkling 2 Mini by 157 rating points13191162+157 rating
  • OmniDocBench v1.6Inkling 2 Mini by 14.992.877.9+14.9
  • JobBenchInkling 2 Mini by 13.149.636.5+13.1
  • Chartography originalInkling 2 Mini by 12.422.19.7+12.4
  • Vals Finance IndexInkling 2 Mini by 9.150.141+9.1
  • GDP (Surge AI)Inkling 2 Mini by 8.816.88+8.8
  • DeepSearchQA (F1)Inkling 2 Mini by 7.685.377.7+7.6
  • GraphWalks BFS 256KInkling 2 Mini by 6.469.763.3+6.4
  • HLE-DiamondInkling 2 Mini by 6.415.28.8+6.4
  • ZeroBench (main) originalInkling 2 Mini by 61711+6
  • Terminal-Bench 4.0Inkling 2 Mini by 5.65.60+5.6
  • CritPt v2 FullNemotron 3.5 Super VL by 5.531.436.9+5.5
  • CritPt v2 HardInkling 2 Mini by 3.51713.5+3.5
  • ComplexConstraints v1.1Inkling 2 Mini by 3.333.430.1+3.3
  • MMMU Pro Standard 10Inkling 2 Mini by 2.674.972.3+2.6
  • StrongREJECT (none)Nemotron 3.5 Super VL by 1.697.899.4+1.6

Questions people ask

Which is better, Inkling 2 Mini or Nemotron 3.5 Super VL?

Inkling 2 Mini wins all four areas where both have results: coding, agents, reasoning and long documents. Nemotron 3.5 Super VL wins none.

Which is better for coding?

Inkling 2 Mini. It wins 3 of the 3 coding tests both models report; Nemotron 3.5 Super VL wins none.

Which is cheaper?

Inkling 2 Mini costs $0.15 per million input tokens and $0.50 per million output tokens; Nemotron 3.5 Super VL costs $0.15 and $0.50. For a million tokens read plus a million written, they cost about the same.

How do you compare the two?

We use the 33 benchmark tests both models have published scores on. The verdict counts the 8 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 25 are listed but not counted, because several are summary scores or repeat a test. Each score is the one shown on the model's own page.