DeepSeek-V4-Flash vs Gemini 3.5 Flash
Wins 2 of 7 areas
Agents · Long documents
Wins 4 of 7 areas
Coding · Reasoning · Facts · Math
Gemini 3.5 Flash wins more areas, narrowly.DeepSeek-V4-Flash is cheaper and better at agents.
Scores updated · 43 tests both models report · How we compare
Where each one wins
Tests won in each of the seven 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 software06Gemini 3.5 Flash6 of 7 tests · 1 tie
- MathCompetition and research-level math03Gemini 3.5 Flash3 of 5 tests · 2 ties
- FactsGetting facts right instead of making them up03Gemini 3.5 Flash3 of 3 tests
- ReasoningHard problems that need careful thinking24Gemini 3.5 Flash4 of 6 tests
- AgentsCarrying out multi-step tasks on its own32DeepSeek-V4-Flash3 of 5 tests
- Long documentsFinding answers in very long texts10DeepSeek-V4-Flash1 of 1 test
- Following instructionsDoing exactly what it is asked11Even1 each
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.
DeepSeek-V4-Flash costs 83% less for the same work.
The biggest differences
The three tests each model wins by the widest margin. Scores are out of 100.
Where Gemini 3.5 Flash pulls ahead
- Short factual questions, answered correctlySimpleQA Verified+32.1points ahead
- Common-sense trick questionsSimpleBench+30.4points ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+29.5points ahead
Where DeepSeek-V4-Flash pulls ahead
- Real work tasks from 44 professionsGDPVal+11.7points ahead
- Customer-service tasks in a simulated bankτ-Bench V3 · Banking+7.2points ahead
- Reasons across sets of long documentsAA-LCR+6.4points ahead
Every test, side by side
All 43 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingGemini 3.5 Flash
- LMArena · WebDevGemini 3.5 Flash by 69 rating points14301499+69 rating
- Terminal-Bench HardGemini 3.5 Flash by 5.335.640.9+5.3
- SciCodeGemini 3.5 Flash by 3.650.353.9+3.6
- LiveBench · CodingGemini 3.5 Flash by 3.27578.2+3.2
- SWE-bench ProGemini 3.5 Flash by 2.552.655.1+2.5
- LiveBench · Agentic CodingGemini 3.5 Flash by 2.246.849+2.2
- Terminal-Bench 2.1tie78.778.7tie
AgentsDeepSeek-V4-Flash
- MCP AtlasGemini 3.5 Flash by 14.66983.6+14.6
- GDPValDeepSeek-V4-Flash by 11.746.935.2+11.7
- AA IT-Bench SREGemini 3.5 Flash by 8.831.540.3+8.8
- τ-Bench V3 · BankingDeepSeek-V4-Flash by 7.239.432.2+7.2
- Terminal-Bench 4.0DeepSeek-V4-Flash by 5.512.16.6+5.5
ReasoningGemini 3.5 Flash
- SimpleBenchGemini 3.5 Flash by 30.446.376.7+30.4
- ARC-AGI-2Gemini 3.5 Flash by 10.761.472.1+10.7
- LiveBench · ReasoningDeepSeek-V4-Flash by 4.686.682+4.6
- Humanity's Last ExamGemini 3.5 Flash by 4.138.642.7+4.1
- CritPtDeepSeek-V4-Flash by 3.516.613.1+3.5
- GPQA DiamondGemini 3.5 Flash by 1.490.892.2+1.4
FactsGemini 3.5 Flash
- SimpleQA VerifiedGemini 3.5 Flash by 32.134.166.2+32.1
- AA-Omniscience · Non-hallucinationGemini 3.5 Flash by 29.58.337.8+29.5
- AA-Omniscience · AccuracyGemini 3.5 Flash by 1140.451.4+11
MathGemini 3.5 Flash
- FrontierMath Tiers 1-3 (v2)Gemini 3.5 Flash by 5.357.562.8+5.3
- FrontierMath Tier 4Gemini 3.5 Flash by 2.424.426.8+2.4
- LiveBench · MathematicsGemini 3.5 Flash by 1.486.888.2+1.4
- AIME 2026tie95.895tie
- HMMT Feb. 2026tie94.895.5tie
Long documentsDeepSeek-V4-Flash
- AA-LCRDeepSeek-V4-Flash by 6.479.773.3+6.4
Following instructionsEven
- LiveBench · Instruction FollowingGemini 3.5 Flash by 10.165.575.6+10.1
- IFBenchDeepSeek-V4-Flash by 2.979.276.3+2.9
Other results14 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-OmniscienceGemini 3.5 Flash by 35.5-14.321.2+35.5
- GDPval-AA (Elo)Gemini 3.5 Flash by 261 rating points13951656+261 rating
- Terminal-Bench 2.0Gemini 3.5 Flash by 19.356.976.2+19.3
- DeepSWEDeepSeek-V4-Flash by 17.454.437+17.4
- DeepSWE 1.1DeepSeek-V4-Flash by 17.454.437+17.4
- GDPval-AA v2Gemini 3.5 Flash by 160 rating points11891349+160 rating
- AA Agentic IndexDeepSeek-V4-Flash by 14.441.727.3+14.4
- livebench_data_analysisDeepSeek-V4-Flash by 14.479.364.9+14.4
- ToolathlonDeepSeek-V4-Flash by 13.870.356.5+13.8
- livebench_languageGemini 3.5 Flash by 5.479.284.6+5.4
- ARC-AGI-1Gemini 3.5 Flash by 3.58992.5+3.5
- AA IntelligenceDeepSeek-V4-Flash by 1.734.332.6+1.7
- Artificial Analysis Coding IndexGemini 3.5 Flash by 169.170.1+1
- τ²-Bench Telecom (AA run)tie9595.3tie
Questions people ask
Which is better, DeepSeek-V4-Flash or Gemini 3.5 Flash?
Gemini 3.5 Flash wins four of the seven areas where both have results: coding, reasoning, facts and math. DeepSeek-V4-Flash wins agents and long documents, and costs 83% less. They are level on following instructions.
Which is better for coding?
Gemini 3.5 Flash. It wins 6 of the 7 coding tests both models report; DeepSeek-V4-Flash wins none, and 1 is a tie.
Which is cheaper?
DeepSeek-V4-Flash costs $0.44 per million input tokens and $1.32 per million output tokens; Gemini 3.5 Flash costs $1.50 and $9.00. That makes DeepSeek-V4-Flash about 83% cheaper for the same work.
How do you compare the two?
We use the 43 benchmark tests both models have published scores on. The verdict counts the 29 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 14 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.