DeepSeek-R1 vs Gemini 2.5 Pro
Wins 0 of 6 areas
—
Wins 6 of 6 areas
Coding · Agents · Reasoning · Facts · Long documents · Following instructions
Gemini 2.5 Pro is the stronger all-rounder.DeepSeek-R1 is cheaper.
Scores updated · 43 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 software04Gemini 2.5 Pro4 of 4 tests
- ReasoningHard problems that need careful thinking03Gemini 2.5 Pro3 of 3 tests
- FactsGetting facts right instead of making them up02Gemini 2.5 Pro2 of 3 tests · 1 tie
- Following instructionsDoing exactly what it is asked02Gemini 2.5 Pro2 of 2 tests
- AgentsCarrying out multi-step tasks on its own01Gemini 2.5 Pro1 of 3 tests · 2 ties
- Long documentsFinding answers in very long texts01Gemini 2.5 Pro1 of 2 tests · 1 tie
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-R1 costs 47% 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 Gemini 2.5 Pro pulls ahead
- Hard command-line tasks in a real terminalTerminal-Bench Hard+20.4points ahead
- Fixes real GitHub issues in Python projectsSWE-bench Verified+14points ahead
- Keeps track of context across a multi-turn chatMulti-Challenge+12.9points ahead
Where DeepSeek-R1 pulls ahead
No clear win on a test scored out of 100.
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 2.5 Pro
- Terminal-Bench HardGemini 2.5 Pro by 20.46.126.5+20.4
- SWE-bench VerifiedGemini 2.5 Pro by 1449.263.2+14
- Terminal-Bench 2.1Gemini 2.5 Pro by 9.419.128.5+9.4
- SciCodeGemini 2.5 Pro by 838.346.3+8
AgentsGemini 2.5 Pro
- τ-Bench V3 · BankingGemini 2.5 Pro by 3.36.49.7+3.3
- GDPValtie00tie
- Terminal-Bench 4.0tie00tie
ReasoningGemini 2.5 Pro
- GPQA DiamondGemini 2.5 Pro by 12.870.883.6+12.8
- Humanity's Last ExamGemini 2.5 Pro by 9.58.518+9.5
- CritPtGemini 2.5 Pro by 20.62.6+2
FactsGemini 2.5 Pro
- AA-Omniscience · AccuracyGemini 2.5 Pro by 8.630.539+8.6
- Vectara HHEM hallucination ratelower is betterGemini 2.5 Pro by 4.311.37+4.3
- AA-Omniscience · Non-hallucinationtie9.79.1tie
Long documentsGemini 2.5 Pro
- AA-LCRGemini 2.5 Pro by 11.357.769+11.3
- LongBench v2tie58.358.8tie
Following instructionsGemini 2.5 Pro
- Multi-ChallengeGemini 2.5 Pro by 12.940.753.6+12.9
- IFBenchGemini 2.5 Pro by 9.73948.7+9.7
Other results26 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.
- τ²-Bench Telecom (AA run)Gemini 2.5 Pro by 42.711.454.1+42.7
- OpenAI-MRCR (128k)Gemini 2.5 Pro by 4135.876.8+41
- HMMT Feb. 2025Gemini 2.5 Pro by 39.141.780.8+39.1
- Aider-PolyglotGemini 2.5 Pro by 23.253.376.5+23.2
- Artificial Analysis Coding IndexGemini 2.5 Pro by 22.124.646.7+22.1
- ARC-AGI-1Gemini 2.5 Pro by 21.215.837+21.2
- LiveCodeBench (24/8~25/5)Gemini 2.5 Pro by 21.255.977.1+21.2
- AIR-Bench 2024Gemini 2.5 Pro by 20.752.973.6+20.7
- SimpleQAGemini 2.5 Pro by 20.730.150.8+20.7
- usamo_2025Gemini 2.5 Pro by 19.64.824.4+19.6
- HarmBenchGemini 2.5 Pro by 17.547.965.4+17.5
- AA-OmniscienceGemini 2.5 Pro by 15.9-32.2-16.4+15.9
- AIME 2025Gemini 2.5 Pro by 137083+13
- ZebraLogicGemini 2.5 Pro by 12.978.791.6+12.9
- AIME 2024Gemini 2.5 Pro by 12.279.892+12.2
- LiveCodeBenchGemini 2.5 Pro by 10.763.574.2+10.7
- vectara_factual_consistencyGemini 2.5 Pro by 4.388.793+4.3
- XSTestGemini 2.5 Pro by 4.394.498.7+4.3
- AA IntelligenceGemini 2.5 Pro by 3.611.415+3.6
- anthropic_red_teamGemini 2.5 Pro by 2.397.299.5+2.3
- vectara_answer_rateGemini 2.5 Pro by 2.19799.1+2.1
- MMLU-ProGemini 2.5 Pro by 28486+2
- MATH-500 (EM)Gemini 2.5 Pro by 1.597.398.8+1.5
- vectara_avg_summary_lengthGemini 2.5 Pro by 12.9 rating points93.5106.4+12.9 rating
- simple_safety_testsDeepSeek-R1 by 19897+1
- bbqtie96.696.4tie
Questions people ask
Which is better, DeepSeek-R1 or Gemini 2.5 Pro?
Gemini 2.5 Pro wins all six areas where both have results: coding, agents, reasoning, facts, long documents and following instructions. DeepSeek-R1 wins none, but costs 47% less.
Which is better for coding?
Gemini 2.5 Pro. It wins 4 of the 4 coding tests both models report; DeepSeek-R1 wins none.
Which is cheaper?
DeepSeek-R1 costs $2.00 per million input tokens and $4.00 per million output tokens; Gemini 2.5 Pro costs $1.25 and $10.00. That makes DeepSeek-R1 about 47% 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 17 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 26 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.