DeepSeek-V4-Pro vs Gemma 4 31B
Wins 5 of 7 areas
Coding · Agents · Reasoning · Math · Long documents
Wins 0 of 7 areas
—
DeepSeek-V4-Pro is the stronger all-rounder.Gemma 4 31B is cheaper.
Scores updated · 40 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 software80DeepSeek-V4-Pro8 of 8 tests
- AgentsCarrying out multi-step tasks on its own50DeepSeek-V4-Pro5 of 5 tests
- ReasoningHard problems that need careful thinking30DeepSeek-V4-Pro3 of 3 tests
- MathCompetition and research-level math30DeepSeek-V4-Pro3 of 3 tests
- Long documentsFinding answers in very long texts10DeepSeek-V4-Pro1 of 1 test
- FactsGetting facts right instead of making them up22Even2 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.
Gemma 4 31B costs 57% 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-Pro pulls ahead
- Short factual questions, answered correctlySimpleQA Verified+35.8points ahead
- Fixes real GitHub issues in Python projectsSWE-bench Verified+28.6points ahead
- Real work tasks from 44 professionsGDPVal+26.8points ahead
Where Gemma 4 31B pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+9.1points ahead
Every test, side by side
All 40 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingDeepSeek-V4-Pro
- SWE-bench VerifiedDeepSeek-V4-Pro by 28.680.652+28.6
- SWE-bench MultilingualDeepSeek-V4-Pro by 24.576.251.7+24.5
- Terminal-Bench 2.1DeepSeek-V4-Pro by 20.66443.4+20.6
- SWE-bench ProDeepSeek-V4-Pro by 19.755.435.7+19.7
- LiveCodeBench v6DeepSeek-V4-Pro by 12.592.580+12.5
- LMArena · WebDevDeepSeek-V4-Pro by 99 rating points14641365+99 rating
- Terminal-Bench HardDeepSeek-V4-Pro by 9.846.236.4+9.8
- SciCodeDeepSeek-V4-Pro by 5.350.845.5+5.3
AgentsDeepSeek-V4-Pro
- GDPValDeepSeek-V4-Pro by 26.832.96.1+26.8
- MCP AtlasDeepSeek-V4-Pro by 16.473.657.2+16.4
- τ-Bench V3 · BankingDeepSeek-V4-Pro by 15.330.114.8+15.3
- Terminal-Bench 4.0DeepSeek-V4-Pro by 14.614.60+14.6
- AA IT-Bench SREDeepSeek-V4-Pro by 138.337.3+1
ReasoningDeepSeek-V4-Pro
- Humanity's Last ExamDeepSeek-V4-Pro by 13.937.523.6+13.9
- CritPtDeepSeek-V4-Pro by 11.512.91.4+11.5
- GPQA DiamondDeepSeek-V4-Pro by 3.188.885.7+3.1
FactsEven
- SimpleQA VerifiedDeepSeek-V4-Pro by 35.846.210.4+35.8
- AA-Omniscience · AccuracyDeepSeek-V4-Pro by 234320+23
- AA-Omniscience · Non-hallucinationGemma 4 31B by 9.15.915+9.1
- Vectara HHEM hallucination ratelower is betterGemma 4 31B by 1.28.67.4+1.2
MathDeepSeek-V4-Pro
- HMMT Feb. 2026DeepSeek-V4-Pro by 1895.277.2+18
- IMOAnswerBenchDeepSeek-V4-Pro by 15.389.874.5+15.3
- AIME 2026DeepSeek-V4-Pro by 7.596.789.2+7.5
Other results15 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-OmniscienceDeepSeek-V4-Pro by 37.2-10.7-47.9+37.2
- τ²-Bench Telecom (AA run)DeepSeek-V4-Pro by 36.396.259.9+36.3
- HLE (with tools)DeepSeek-V4-Pro by 33.56026.5+33.5
- Tool-DecathlonDeepSeek-V4-Pro by 31.652.821.2+31.6
- Terminal-Bench 2.0DeepSeek-V4-Pro by 2567.942.9+25
- NL2RepoDeepSeek-V4-Pro by 2338.515.5+23
- AA Agentic IndexDeepSeek-V4-Pro by 2127.76.7+21
- Artificial Analysis Coding IndexDeepSeek-V4-Pro by 1659.443.4+16
- AA IntelligenceDeepSeek-V4-Pro by 15.730.414.7+15.7
- vectara_avg_summary_lengthDeepSeek-V4-Pro by 78 rating points153.875.8+78 rating
- HMMT Nov. 2025DeepSeek-V4-Pro by 6.994.487.5+6.9
- GPQA (unspecified)DeepSeek-V4-Pro by 3.587.884.3+3.5
- vectara_answer_rateGemma 4 31B by 2.897.2100+2.8
- MMLU-ProDeepSeek-V4-Pro by 2.387.585.2+2.3
- vectara_factual_consistencyGemma 4 31B by 1.291.492.6+1.2
Questions people ask
Which is better, DeepSeek-V4-Pro or Gemma 4 31B?
DeepSeek-V4-Pro wins five of the seven areas where both have results: coding, agents, reasoning, math and long documents. Gemma 4 31B wins none, but costs 57% less. They are level on facts and following instructions.
Which is better for coding?
DeepSeek-V4-Pro. It wins 8 of the 8 coding tests both models report; Gemma 4 31B wins none.
Which is cheaper?
DeepSeek-V4-Pro costs $0.43 per million input tokens and $0.87 per million output tokens; Gemma 4 31B costs $0.17 and $0.40. That makes Gemma 4 31B about 57% cheaper for the same work.
How do you compare the two?
We use the 40 benchmark tests both models have published scores on. The verdict counts the 25 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 15 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.