DeepSeek-V4-Pro vs Gemini 3 Pro
Wins 4 of 7 areas
Coding · Agents · Math · Following instructions
Wins 3 of 7 areas
Reasoning · Facts · Long documents
DeepSeek-V4-Pro wins more areas, narrowly.Gemini 3 Pro is better at reasoning.
Scores updated · 42 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 software61DeepSeek-V4-Pro6 of 7 tests
- MathCompetition and research-level math30DeepSeek-V4-Pro3 of 3 tests
- AgentsCarrying out multi-step tasks on its own31DeepSeek-V4-Pro3 of 4 tests
- Following instructionsDoing exactly what it is asked10DeepSeek-V4-Pro1 of 1 test
- ReasoningHard problems that need careful thinking13Gemini 3 Pro3 of 4 tests
- FactsGetting facts right instead of making them up13Gemini 3 Pro3 of 4 tests
- Long documentsFinding answers in very long texts01Gemini 3 Pro1 of 1 test
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.
DeepSeek-V4-Pro costs 91% less for the same work.
The biggest differences
The three tests each model wins by the widest margin. Scores are out of 100.
Where DeepSeek-V4-Pro pulls ahead
- Finds hard-to-locate facts by browsing the webBrowseComp+24.2points ahead
- Multi-step tasks using many real software toolsMCP Atlas+19.5points ahead
- Long, multi-file coding tasks in real codebasesSWE-bench Pro+12.1points ahead
Where Gemini 3 Pro pulls ahead
- Short factual questions, answered correctlySimpleQA Verified+26.7points ahead
- Common-sense trick questionsSimpleBench+25.5points ahead
- Answers hard knowledge questions correctlyAA-Omniscience · Accuracy+12.8points ahead
Every test, side by side
All 42 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingDeepSeek-V4-Pro
- SWE-bench ProDeepSeek-V4-Pro by 12.155.443.3+12.1
- SWE-bench MultilingualDeepSeek-V4-Pro by 11.276.265+11.2
- SciCodeGemini 3 Pro by 5.350.856.1+5.3
- LiveCodeBench v6DeepSeek-V4-Pro by 5.192.587.4+5.1
- Terminal-Bench HardDeepSeek-V4-Pro by 4.546.241.7+4.5
- SWE-bench VerifiedDeepSeek-V4-Pro by 4.480.676.2+4.4
- LMArena · WebDevDeepSeek-V4-Pro by 24 rating points14641440+24 rating
AgentsDeepSeek-V4-Pro
- BrowseCompDeepSeek-V4-Pro by 24.283.459.2+24.2
- MCP AtlasDeepSeek-V4-Pro by 19.573.654.1+19.5
- AA ApexAgentsDeepSeek-V4-Pro by 5.924.318.4+5.9
- GDPValGemini 3 Pro by 1.332.934.2+1.3
ReasoningGemini 3 Pro
- SimpleBenchGemini 3 Pro by 25.550.976.4+25.5
- CritPtDeepSeek-V4-Pro by 3.812.99.1+3.8
- Humanity's Last ExamGemini 3 Pro by 2.237.539.7+2.2
- GPQA DiamondGemini 3 Pro by 288.890.8+2
FactsGemini 3 Pro
- SimpleQA VerifiedGemini 3 Pro by 26.746.272.9+26.7
- AA-Omniscience · AccuracyGemini 3 Pro by 12.84355.8+12.8
- Vectara HHEM hallucination ratelower is betterDeepSeek-V4-Pro by 58.613.6+5
- AA-Omniscience · Non-hallucinationGemini 3 Pro by 2.65.98.5+2.6
MathDeepSeek-V4-Pro
- HMMT Feb. 2026DeepSeek-V4-Pro by 8.895.286.4+8.8
- IMOAnswerBenchDeepSeek-V4-Pro by 6.789.883.1+6.7
- AIME 2026DeepSeek-V4-Pro by 596.791.7+5
Following instructionsDeepSeek-V4-Pro
- IFBenchDeepSeek-V4-Pro by 6.176.570.4+6.1
Other results18 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.
- MathArena Apex (Pass@1)DeepSeek-V4-Pro by 66.890.223.4+66.8
- GDPval-AA (Elo)DeepSeek-V4-Pro by 359 rating points15541195+359 rating
- AA-OmniscienceGemini 3 Pro by 26-10.715.3+26
- AA Agentic IndexGemini 3 Pro by 24.327.752+24.3
- ToolathlonDeepSeek-V4-Pro by 15.451.836.4+15.4
- HLE (with tools)DeepSeek-V4-Pro by 14.26045.8+14.2
- SimpleQAGemini 3 Pro by 14.257.972.1+14.2
- Terminal-Bench 2.0DeepSeek-V4-Pro by 13.767.954.2+13.7
- Artificial Analysis Coding IndexDeepSeek-V4-Pro by 12.959.446.5+12.9
- τ²-Bench Telecom (AA run)DeepSeek-V4-Pro by 9.196.287.1+9.1
- vectara_avg_summary_lengthDeepSeek-V4-Pro by 51.9 rating points153.8101.9+51.9 rating
- vectara_factual_consistencyDeepSeek-V4-Pro by 591.486.4+5
- MMLU-ProGemini 3 Pro by 2.687.590.1+2.6
- AA IntelligenceDeepSeek-V4-Pro by 2.430.428+2.4
- vectara_answer_rateGemini 3 Pro by 2.297.299.4+2.2
- LiveCodeBenchDeepSeek-V4-Pro by 1.593.592+1.5
- HMMT Nov. 2025DeepSeek-V4-Pro by 1.194.493.3+1.1
- LiveBenchtie73.673.4tie
Questions people ask
Which is better, DeepSeek-V4-Pro or Gemini 3 Pro?
DeepSeek-V4-Pro wins four of the seven areas where both have results: coding, agents, math and following instructions. Gemini 3 Pro wins reasoning, facts and long documents.
Which is better for coding?
DeepSeek-V4-Pro. It wins 6 of the 7 coding tests both models report; Gemini 3 Pro wins 1.
Which is cheaper?
DeepSeek-V4-Pro costs $0.43 per million input tokens and $0.87 per million output tokens; Gemini 3 Pro costs $2.00 and $12.00. That makes DeepSeek-V4-Pro about 91% cheaper for the same work.
How do you compare the two?
We use the 42 benchmark tests both models have published scores on. The verdict counts the 24 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 18 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.