Gemma 4 26B A4B vs Nemotron 3.5 Lightning
Wins 2 of 6 areas
Reasoning · Long documents
Wins 1 of 6 areas
Agents
Gemma 4 26B A4B wins more areas, narrowly.Nemotron 3.5 Lightning is cheaper and better at agents.
Scores updated · 36 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.
- ReasoningHard problems that need careful thinking20Gemma 4 26B A4B2 of 3 tests · 1 tie
- Long documentsFinding answers in very long texts10Gemma 4 26B A4B1 of 1 test
- AgentsCarrying out multi-step tasks on its own12Nemotron 3.5 Lightning2 of 3 tests
- CodingWriting and fixing software22Even2 each
- FactsGetting facts right instead of making them up11Even1 each
- Following instructionsDoing exactly what it is asked00Even0 each · 1 tie
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.
Nemotron 3.5 Lightning costs 51% less for the same work.
The biggest differences
The three tests each model wins by the widest margin. Scores are out of 100.
Where Gemma 4 26B A4B pulls ahead
- Command-line tasks in a real terminalTerminal-Bench 2.1+14.7points ahead
- Very hard expert questions across many subjectsHumanity's Last Exam+8.7points ahead
- Code for real scientific research problemsSciCode+7.9points ahead
Where Nemotron 3.5 Lightning pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+48.8points ahead
- Fixes real GitHub issues in Python projectsSWE-bench Verified+34.2points ahead
- Fixes real GitHub issues in many programming languagesSWE-bench Multilingual+22points ahead
Every test, side by side
All 36 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingEven
- SWE-bench VerifiedNemotron 3.5 Lightning by 34.217.451.6+34.2
- SWE-bench MultilingualNemotron 3.5 Lightning by 2217.339.3+22
- Terminal-Bench 2.1Gemma 4 26B A4B by 14.73924.3+14.7
- SciCodeGemma 4 26B A4B by 7.94032.1+7.9
AgentsNemotron 3.5 Lightning
- BrowseCompNemotron 3.5 Lightning by 10.726.337+10.7
- GDPValNemotron 3.5 Lightning by 3.63.47+3.6
- τ-Bench V3 · BankingGemma 4 26B A4B by 3.1128.9+3.1
ReasoningGemma 4 26B A4B
- Humanity's Last ExamGemma 4 26B A4B by 8.719.310.6+8.7
- GPQA DiamondGemma 4 26B A4B by 4.979.274.3+4.9
- CritPttie00tie
FactsEven
- AA-Omniscience · Non-hallucinationNemotron 3.5 Lightning by 48.813.662.4+48.8
- AA-Omniscience · AccuracyGemma 4 26B A4B by 4.719.114.4+4.7
Other results22 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.
- Minerva MathNemotron 3.5 Lightning by 39.143.782.8+39.1
- AA-OmniscienceNemotron 3.5 Lightning by 33.1-50.8-17.7+33.1
- HumanEvalNemotron 3.5 Lightning by 27.45077.4+27.4
- Artificial Analysis Coding IndexGemma 4 26B A4B by 12.539.326.8+12.5
- PinchBenchNemotron 3.5 Lightning by 10.774.785.4+10.7
- MBPPNemotron 3.5 Lightning by 10.268.478.6+10.2
- RULER 256KGemma 4 26B A4B by 8.885.776.9+8.8
- HLE (text-only, no tools)Gemma 4 26B A4B by 5.717.411.7+5.7
- AA Agentic IndexGemma 4 26B A4B by 4.9116.1+4.9
- τ³-Bench BankingGemma 4 26B A4B by 4.7149.3+4.7
- AA IntelligenceGemma 4 26B A4B by 3.816.712.9+3.8
- RULER 1MGemma 4 26B A4B by 3.372.969.6+3.3
- GDPval-AA v2Nemotron 3.5 Lightning by 25 rating points807832+25 rating
- PIQA (acc-norm)Nemotron 3.5 Lightning by 1.483.885.2+1.4
- OpenBookQAGemma 4 26B A4B by 1.248.847.6+1.2
- PIQA (Acc.)Nemotron 3.5 Lightning by 1.282.283.3+1.2
- WinoGrandetie79.180tie
- MMLUtie77.878.6tie
- Global-MMLU-Lite (5-shot) Averagetie74.875.5tie
- MMLU-Protie82.681.9tie
- HellaSwagtie85.385.5tie
- ARC-Challengetie92.892.7tie
Questions people ask
Which is better, Gemma 4 26B A4B or Nemotron 3.5 Lightning?
Gemma 4 26B A4B wins two of the six areas where both have results: reasoning and long documents. Nemotron 3.5 Lightning wins agents, and costs 51% less. They are level on coding, facts and following instructions.
Which is better for coding?
Neither. They win 2 coding tests each of the 4 both models report.
Which is cheaper?
Gemma 4 26B A4B costs $0.13 per million input tokens and $0.40 per million output tokens; Nemotron 3.5 Lightning costs $0.06 and $0.20. That makes Nemotron 3.5 Lightning about 51% cheaper for the same work.
How do you compare the two?
We use the 36 benchmark tests both models have published scores on. The verdict counts the 14 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 22 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.