DeepSeek-V4-Flash vs Nemotron 3.5 Lightning
Wins 5 of 6 areas
Coding · Agents · Reasoning · Long documents · Following instructions
Wins 0 of 6 areas
—
DeepSeek-V4-Flash is the stronger all-rounder.Nemotron 3.5 Lightning is cheaper.
Scores updated · 23 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 software40DeepSeek-V4-Flash4 of 4 tests
- AgentsCarrying out multi-step tasks on its own40DeepSeek-V4-Flash4 of 4 tests
- ReasoningHard problems that need careful thinking30DeepSeek-V4-Flash3 of 3 tests
- Long documentsFinding answers in very long texts10DeepSeek-V4-Flash1 of 1 test
- Following instructionsDoing exactly what it is asked10DeepSeek-V4-Flash1 of 1 test
- FactsGetting facts right instead of making them up11Even1 each
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 85% less for the same work.
The biggest differences
The tests each model wins by the widest margin, up to three each. Scores are out of 100.
Where DeepSeek-V4-Flash pulls ahead
- Command-line tasks in a real terminalTerminal-Bench 2.1+54.4points ahead
- Real work tasks from 44 professionsGDPVal+39.9points ahead
- Finds hard-to-locate facts by browsing the webBrowseComp+36.2points ahead
Where Nemotron 3.5 Lightning pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+54.1points ahead
Every test, side by side
All 23 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingDeepSeek-V4-Flash
- Terminal-Bench 2.1DeepSeek-V4-Flash by 54.478.724.3+54.4
- SWE-bench MultilingualDeepSeek-V4-Flash by 3473.339.3+34
- SWE-bench VerifiedDeepSeek-V4-Flash by 27.47951.6+27.4
- SciCodeDeepSeek-V4-Flash by 18.250.332.1+18.2
AgentsDeepSeek-V4-Flash
- GDPValDeepSeek-V4-Flash by 39.946.97+39.9
- BrowseCompDeepSeek-V4-Flash by 36.273.237+36.2
- τ-Bench V3 · BankingDeepSeek-V4-Flash by 30.539.48.9+30.5
- Terminal-Bench 4.0DeepSeek-V4-Flash by 11.612.10.5+11.6
ReasoningDeepSeek-V4-Flash
- Humanity's Last ExamDeepSeek-V4-Flash by 2838.610.6+28
- CritPtDeepSeek-V4-Flash by 16.616.60+16.6
- GPQA DiamondDeepSeek-V4-Flash by 16.590.874.3+16.5
FactsEven
- AA-Omniscience · Non-hallucinationNemotron 3.5 Lightning by 54.18.362.4+54.1
- AA-Omniscience · AccuracyDeepSeek-V4-Flash by 2640.414.4+26
Long documentsDeepSeek-V4-Flash
- AA-LCRDeepSeek-V4-Flash by 19.479.760.3+19.4
Following instructionsDeepSeek-V4-Flash
- IFBenchDeepSeek-V4-Flash by 7.379.271.9+7.3
Other results8 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.
- Artificial Analysis Coding IndexDeepSeek-V4-Flash by 42.369.126.8+42.3
- GDPval-AA v2DeepSeek-V4-Flash by 357 rating points1189832+357 rating
- AA Agentic IndexDeepSeek-V4-Flash by 35.641.76.1+35.6
- AA IntelligenceDeepSeek-V4-Flash by 21.434.312.9+21.4
- τ³-Bench BankingDeepSeek-V4-Flash by 13.622.99.3+13.6
- PinchBenchDeepSeek-V4-Flash by 5.991.385.4+5.9
- MMLU-ProDeepSeek-V4-Flash by 4.386.281.9+4.3
- AA-OmniscienceDeepSeek-V4-Flash by 3.4-14.3-17.7+3.4
Questions people ask
Which is better, DeepSeek-V4-Flash or Nemotron 3.5 Lightning?
DeepSeek-V4-Flash wins five of the six areas where both have results: coding, agents, reasoning, long documents and following instructions. Nemotron 3.5 Lightning wins none, but costs 85% less. They are level on facts.
Which is better for coding?
DeepSeek-V4-Flash. It wins 4 of the 4 coding tests both models report; Nemotron 3.5 Lightning wins none.
Which is cheaper?
DeepSeek-V4-Flash costs $0.44 per million input tokens and $1.32 per million output tokens; Nemotron 3.5 Lightning costs $0.06 and $0.20. That makes Nemotron 3.5 Lightning about 85% cheaper for the same work.
How do you compare the two?
We use the 23 benchmark tests both models have published scores on. The verdict counts the 15 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 8 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.