DeepSeek-V3.1-Terminus vs Gemini 2.5 Pro
Wins 3 of 6 areas
Coding · Agents · Following instructions
Wins 1 of 6 areas
Reasoning
DeepSeek-V3.1-Terminus wins more areas, narrowly.Gemini 2.5 Pro is better at reasoning.
Scores updated · 21 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.
- AgentsCarrying out multi-step tasks on its own20DeepSeek-V3.1-Terminus2 of 3 tests · 1 tie
- CodingWriting and fixing software21DeepSeek-V3.1-Terminus2 of 3 tests
- Following instructionsDoing exactly what it is asked10DeepSeek-V3.1-Terminus1 of 2 tests · 1 tie
- ReasoningHard problems that need careful thinking02Gemini 2.5 Pro2 of 3 tests · 1 tie
- FactsGetting facts right instead of making them up11Even1 each
- Long documentsFinding answers in very long texts00Even0 each · 1 tie
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-V3.1-Terminus costs 89% 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-V3.1-Terminus pulls ahead
- Command-line tasks in a real terminalTerminal-Bench 2.1+16.4points ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+16.1points ahead
- Customer-service tasks in a simulated bankτ-Bench V3 · Banking+11.3points ahead
Where Gemini 2.5 Pro pulls ahead
- Answers hard knowledge questions correctlyAA-Omniscience · Accuracy+11.4points ahead
- Code for real scientific research problemsSciCode+8.3points ahead
- Graduate-level biology, physics and chemistry questionsGPQA Diamond+4.4points ahead
Every test, side by side
All 21 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingDeepSeek-V3.1-Terminus
- Terminal-Bench 2.1DeepSeek-V3.1-Terminus by 16.444.928.5+16.4
- SciCodeGemini 2.5 Pro by 8.33846.3+8.3
- Terminal-Bench HardDeepSeek-V3.1-Terminus by 3.830.326.5+3.8
AgentsDeepSeek-V3.1-Terminus
- τ-Bench V3 · BankingDeepSeek-V3.1-Terminus by 11.3219.7+11.3
- GDPValDeepSeek-V3.1-Terminus by 10.610.60+10.6
- Terminal-Bench 4.0tie00tie
ReasoningGemini 2.5 Pro
- GPQA DiamondGemini 2.5 Pro by 4.479.283.6+4.4
- Humanity's Last ExamGemini 2.5 Pro by 1.616.418+1.6
- CritPttie1.72.6tie
FactsEven
- AA-Omniscience · Non-hallucinationDeepSeek-V3.1-Terminus by 16.125.29.1+16.1
- AA-Omniscience · AccuracyGemini 2.5 Pro by 11.427.739+11.4
Following instructionsDeepSeek-V3.1-Terminus
- IFBenchDeepSeek-V3.1-Terminus by 8.35748.7+8.3
- Multi-Challengetie54.453.6tie
Other results7 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.
- HMMT 2025DeepSeek-V3.1-Terminus by 20.486.165.7+20.4
- τ²-Bench Telecom (AA run)Gemini 2.5 Pro by 1737.154.1+17
- AA-OmniscienceGemini 2.5 Pro by 10.1-26.4-16.4+10.1
- AA Agentic IndexDeepSeek-V3.1-Terminus by 5.48.93.5+5.4
- Artificial Analysis Coding IndexGemini 2.5 Pro by 3.243.546.7+3.2
- LiveCodeBenchtie74.974.2tie
- AA Intelligencetie14.815tie
Questions people ask
Which is better, DeepSeek-V3.1-Terminus or Gemini 2.5 Pro?
DeepSeek-V3.1-Terminus wins three of the six areas where both have results: coding, agents and following instructions. Gemini 2.5 Pro wins reasoning. They are level on facts and long documents.
Which is better for coding?
DeepSeek-V3.1-Terminus. It wins 2 of the 3 coding tests both models report; Gemini 2.5 Pro wins 1.
Which is cheaper?
DeepSeek-V3.1-Terminus costs $0.27 per million input tokens and $1.00 per million output tokens; Gemini 2.5 Pro costs $1.25 and $10.00. That makes DeepSeek-V3.1-Terminus about 89% cheaper for the same work.
How do you compare the two?
We use the 21 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 7 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.