LongCat Flash Lite vs Nemotron 3 Super 120B A12B
Wins 1 of 6 areas
Agents
Wins 5 of 6 areas
Coding · Reasoning · Facts · Long documents · Following instructions
Nemotron 3 Super 120B A12B is the stronger all-rounder.LongCat Flash Lite is better at agents.
Scores updated · 22 tests both models report · How we compare
Where each one wins
Tests won in each of the six 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 software03Nemotron 3 Super 120B A12B3 of 4 tests · 1 tie
- ReasoningHard problems that need careful thinking03Nemotron 3 Super 120B A12B3 of 3 tests
- FactsGetting facts right instead of making them up02Nemotron 3 Super 120B A12B2 of 2 tests
- Long documentsFinding answers in very long texts01Nemotron 3 Super 120B A12B1 of 1 test
- Following instructionsDoing exactly what it is asked01Nemotron 3 Super 120B A12B1 of 1 test
- AgentsCarrying out multi-step tasks on its own10LongCat Flash Lite1 of 1 test
Agents, long documents 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 tests each model wins by the widest margin, up to three each. Scores are out of 100.
Where Nemotron 3 Super 120B A12B pulls ahead
- Reasons across sets of long documentsAA-LCR+37.4points ahead
- Follows unfamiliar, precisely checkable instructionsIFBench+28.4points ahead
- Hard command-line tasks in a real terminalTerminal-Bench Hard+18.2points ahead
Where LongCat Flash Lite pulls ahead
No clear win on a test scored out of 100.
Every test, side by side
All 22 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingNemotron 3 Super 120B A12B
- Terminal-Bench HardNemotron 3 Super 120B A12B by 18.210.628.8+18.2
- SciCodeNemotron 3 Super 120B A12B by 7.828.436.2+7.8
- SWE-bench MultilingualNemotron 3 Super 120B A12B by 7.738.145.8+7.7
- SWE-bench Verifiedtie54.453.7tie
ReasoningNemotron 3 Super 120B A12B
- GPQA DiamondNemotron 3 Super 120B A12B by 16.463.680+16.4
- Humanity's Last ExamNemotron 3 Super 120B A12B by 155.820.8+15
- CritPtNemotron 3 Super 120B A12B by 3.103.1+3.1
FactsNemotron 3 Super 120B A12B
- AA-Omniscience · AccuracyNemotron 3 Super 120B A12B by 1113.324.3+11
- AA-Omniscience · Non-hallucinationNemotron 3 Super 120B A12B by 9.23.813+9.2
Long documentsNemotron 3 Super 120B A12B
- AA-LCRNemotron 3 Super 120B A12B by 37.428.365.7+37.4
Following instructionsNemotron 3 Super 120B A12B
- IFBenchNemotron 3 Super 120B A12B by 28.443.171.5+28.4
Other results10 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.
- AA Agentic IndexLongCat Flash Lite by 34.738.84.1+34.7
- AA-OmniscienceNemotron 3 Super 120B A12B by 28.6-70.1-41.5+28.6
- AIME 2025Nemotron 3 Super 120B A12B by 2763.290.2+27
- Artificial Analysis Coding IndexNemotron 3 Super 120B A12B by 21.216.537.7+21.2
- τ²-Bench Telecom (AA run)LongCat Flash Lite by 11.779.567.8+11.7
- τ²-Bench (Retail)LongCat Flash Lite by 10.373.162.8+10.3
- Terminal-BenchLongCat Flash Lite by 833.825.8+8
- MMLU-ProNemotron 3 Super 120B A12B by 5.478.383.7+5.4
- Tau2 airlineLongCat Flash Lite by 1.75856.3+1.7
- AA IntelligenceNemotron 3 Super 120B A12B by 1.311.512.8+1.3
Questions people ask
Which is better, LongCat Flash Lite or Nemotron 3 Super 120B A12B?
Nemotron 3 Super 120B A12B wins five of the six areas where both have results: coding, reasoning, facts, long documents and following instructions. LongCat Flash Lite wins agents.
Which is better for coding?
Nemotron 3 Super 120B A12B. It wins 3 of the 4 coding tests both models report; LongCat Flash Lite wins none, and 1 is a tie.
How do you compare the two?
We use the 22 benchmark tests both models have published scores on. The verdict counts the 12 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 10 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.