Gemini 2.5 Pro vs Kimi K2.5
Wins 1 of 7 areas
Facts
Wins 6 of 7 areas
Coding · Agents · Reasoning · Images and charts · Long documents · Following instructions
Kimi K2.5 is the stronger all-rounder.Gemini 2.5 Pro is better at facts.
Scores updated · 48 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 software05Kimi K2.55 of 5 tests
- Images and chartsUnderstanding pictures, charts and video03Kimi K2.53 of 5 tests · 2 ties
- ReasoningHard problems that need careful thinking03Kimi K2.53 of 4 tests · 1 tie
- AgentsCarrying out multi-step tasks on its own03Kimi K2.53 of 3 tests
- Long documentsFinding answers in very long texts02Kimi K2.52 of 2 tests
- Following instructionsDoing exactly what it is asked02Kimi K2.52 of 2 tests
- FactsGetting facts right instead of making them up31Gemini 2.5 Pro3 of 4 tests
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.
Kimi K2.5 costs 68% 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 Kimi K2.5 pulls ahead
- Finds hard-to-locate facts by browsing the webBrowseComp+65points ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+25.2points ahead
- Follows unfamiliar, precisely checkable instructionsIFBench+21.5points ahead
Where Gemini 2.5 Pro pulls ahead
- Short factual questions, answered correctlySimpleQA Verified+21.7points ahead
- Answers hard knowledge questions correctlyAA-Omniscience · Accuracy+3.9points ahead
Every test, side by side
All 48 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingKimi K2.5
- LMArena · WebDevKimi K2.5 by 212 rating points12241436+212 rating
- Terminal-Bench 2.1Kimi K2.5 by 17.228.545.7+17.2
- SWE-bench VerifiedKimi K2.5 by 13.663.276.8+13.6
- Terminal-Bench HardKimi K2.5 by 8.326.534.8+8.3
- SciCodeKimi K2.5 by 2.746.349+2.7
AgentsKimi K2.5
- BrowseCompKimi K2.5 by 659.974.9+65
- GDPValKimi K2.5 by 17.2017.2+17.2
- τ-Bench V3 · BankingKimi K2.5 by 4.59.714.2+4.5
ReasoningKimi K2.5
- Humanity's Last ExamKimi K2.5 by 12.71830.7+12.7
- ARC-AGI-2Kimi K2.5 by 6.94.911.8+6.9
- GPQA DiamondKimi K2.5 by 4.383.687.9+4.3
- CritPttie2.63.1tie
FactsGemini 2.5 Pro
- AA-Omniscience · Non-hallucinationKimi K2.5 by 25.29.134.3+25.2
- SimpleQA VerifiedGemini 2.5 Pro by 21.75634.3+21.7
- Vectara HHEM hallucination ratelower is betterGemini 2.5 Pro by 7.2714.2+7.2
- AA-Omniscience · AccuracyGemini 2.5 Pro by 3.93935.2+3.9
Images and chartsKimi K2.5
- OCRBenchKimi K2.5 by 6.485.992.3+6.4
- MathVistaKimi K2.5 by 6.283.990.1+6.2
- Video-MMEKimi K2.5 by 2.684.887.4+2.6
- LMArena · Visiontie12631269tie
- MMMU-Protie74.975.4tie
Long documentsKimi K2.5
- AA-LCRKimi K2.5 by 96978+9
- LongBench v2Kimi K2.5 by 2.258.861+2.2
Following instructionsKimi K2.5
- IFBenchKimi K2.5 by 21.548.770.2+21.5
- Multi-ChallengeKimi K2.5 by 7.853.661.4+7.8
Other results23 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)Kimi K2.5 by 41.854.195.9+41.8
- τ²-BenchKimi K2.5 by 31.45485.4+31.4
- BrowseComp-ZHKimi K2.5 by 30.132.262.3+30.1
- HMMT 2025Kimi K2.5 by 29.765.795.4+29.7
- ARC-AGI-1Kimi K2.5 by 28.33765.3+28.3
- HLE (with tools)Kimi K2.5 by 21.828.450.2+21.8
- xbench-DeepSearchKimi K2.5 by 20.75676.7+20.7
- AA Agentic IndexKimi K2.5 by 18.23.521.7+18.2
- HMMT Feb. 2025Kimi K2.5 by 14.680.895.4+14.6
- AIME 2025Kimi K2.5 by 13.18396.1+13.1
- MATH-VisionKimi K2.5 by 10.973.384.2+10.9
- HMMT Nov. 2025Kimi K2.5 by 9.28089.2+9.2
- AA-OmniscienceKimi K2.5 by 9-16.4-7.3+9
- AA IntelligenceKimi K2.5 by 8.51523.5+8.5
- vectara_factual_consistencyGemini 2.5 Pro by 7.29385.8+7.2
- vectara_answer_rateGemini 2.5 Pro by 6.999.192.2+6.9
- Global-MMLU-LiteGemini 2.5 Pro by 4.688.684+4.6
- matharena_visual_math_overallKimi K2.5 by 3.477.280.6+3.4
- VideoMMMUKimi K2.5 by 383.686.6+3
- MMLU-ProKimi K2.5 by 1.18687.1+1.1
- vectara_avg_summary_lengthtie106.4112tie
- Artificial Analysis Coding Indextie46.746.8tie
- frontiermath_tier_4_v1tie4.24.2tie
Questions people ask
Which is better, Gemini 2.5 Pro or Kimi K2.5?
Kimi K2.5 wins six of the seven areas where both have results: coding, agents, reasoning, images and charts, long documents and following instructions. Gemini 2.5 Pro wins facts.
Which is better for coding?
Kimi K2.5. It wins 5 of the 5 coding tests both models report; Gemini 2.5 Pro wins none.
Which is cheaper?
Gemini 2.5 Pro costs $1.25 per million input tokens and $10.00 per million output tokens; Kimi K2.5 costs $0.60 and $3.00. That makes Kimi K2.5 about 68% cheaper for the same work.
How do you compare the two?
We use the 48 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 23 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.