Inkling-Small vs NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16
Wins 2 of 5 areas
Coding · Following instructions
Wins 2 of 5 areas
Reasoning · Images and charts
The two are evenly matched.Inkling-Small is better at coding and following instructions; NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16 at reasoning and images and charts.
Scores updated · 30 tests both models report · How we compare
Where each one wins
Tests won in each of the five 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.
- CodingWriting and fixing software31Inkling-Small3 of 5 tests · 1 tie
- Following instructionsDoing exactly what it is asked10Inkling-Small1 of 1 test
- ReasoningHard problems that need careful thinking01NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF161 of 2 tests · 1 tie
- Images and chartsUnderstanding pictures, charts and video01NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF161 of 1 test
- AgentsCarrying out multi-step tasks on its own00Even0 each · 1 tie
Agents, images and charts and following instructions rest on a single test each.
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.
The biggest differences
The three tests each model wins by the widest margin. Scores are out of 100.
Where Inkling-Small pulls ahead
- Follows unfamiliar, precisely checkable instructionsIFBench+11.3points ahead
- Fixes real GitHub issues in Python projectsSWE-bench Verified+9.1points ahead
- Long, multi-file coding tasks in real codebasesSWE-bench Pro+4.1points ahead
Where NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16 pulls ahead
- Command-line tasks in a real terminalTerminal-Bench 2.1+5.6points ahead
- Reasoning about charts from research papersCharXiv (RQ)+3.4points ahead
- Research-level physics problemsCritPt+2.6points ahead
Every test, side by side
All 30 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingInkling-Small
- SWE-bench VerifiedInkling-Small by 9.180.271.1+9.1
- Terminal-Bench 2.1NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16 by 5.655.160.7+5.6
- SWE-bench ProInkling-Small by 4.155.951.8+4.1
- SWE-bench MultilingualInkling-Small by 3.771.467.7+3.7
- LiveCodeBench v6tie88.388.2tie
ReasoningNVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16
- CritPtNVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16 by 2.68.310.9+2.6
- GPQA Diamondtie89.589.5tie
Images and chartsNVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16
- CharXiv (RQ)NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16 by 3.477.480.8+3.4
Following instructionsInkling-Small
- IFBenchInkling-Small by 11.382.270.9+11.3
Other results20 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.
- WebVoyager [offline]NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16 by 72.45.377.7+72.4
- tau2-bench telecomInkling-Small by 16.990.173.2+16.9
- OCRBench v2_enNVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16 by 13.954.268.1+13.9
- tau2-bench airlineNVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16 by 13.862.776.5+13.8
- MMlongBench DocNVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16 by 13.749.162.8+13.7
- MMLU-ProXInkling-Small by 7.18476.9+7.1
- OCRBench v2, ZHNVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16 by 6.852.459.1+6.8
- DeepSWE [offline]NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16 by 6.71824.6+6.7
- BabyVisionNVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16 by 4.924.729.6+4.9
- BrowseComp [offline]NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16 by 47478+4
- τ³-Bench BankingNVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16 by 3.415.518.9+3.4
- GDPval-AA v2Inkling-Small by 27 rating points12381211+27 rating
- tau2-bench retailInkling-Small by 2.671.869.2+2.6
- HLE, full setInkling-Small by 2.431.629.2+2.4
- HLE, text-onlyInkling-Small by 1.533.331.8+1.5
- SciCode, subtaskInkling-Small by 1.248.347.1+1.2
- LMArena v3 proxytie53.652.7tie
- MATH-Visiontie78.878.2tie
- Terminal Bench 4 [offline]tie0.51tie
- WMT24++tie84.684.8tie
Questions people ask
Which is better, Inkling-Small or NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16?
Inkling-Small and NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16 each win two of the five areas where both have results. Inkling-Small wins coding and following instructions; NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16 wins reasoning and images and charts. They are level on agents.
Which is better for coding?
Inkling-Small. It wins 3 of the 5 coding tests both models report; NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16 wins 1, and 1 is a tie.
How do you compare the two?
We use the 30 benchmark tests both models have published scores on. The verdict counts the 10 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 20 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.