Claude Opus 4 vs Gemini 2.5 Pro
Wins 2 of 6 areas
Coding · Following instructions
Wins 3 of 6 areas
Facts · Images and charts · Long documents
Gemini 2.5 Pro wins more areas, narrowly.Claude Opus 4 is better at following instructions.
Scores updated · 47 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.
- Images and chartsUnderstanding pictures, charts and video02Gemini 2.5 Pro2 of 2 tests
- Long documentsFinding answers in very long texts01Gemini 2.5 Pro1 of 2 tests · 1 tie
- FactsGetting facts right instead of making them up01Gemini 2.5 Pro1 of 1 test
- Following instructionsDoing exactly what it is asked20Claude Opus 42 of 2 tests
- CodingWriting and fixing software21Claude Opus 42 of 3 tests
- ReasoningHard problems that need careful thinking22Even2 each
Facts 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.
Gemini 2.5 Pro costs 88% less for the same work.
The biggest differences
The three tests each model wins by the widest margin. Scores are out of 100.
Where Gemini 2.5 Pro pulls ahead
- Code for real scientific research problemsSciCode+6.5points ahead
- College exam questions with charts, maps and diagramsMMMU+5.9points ahead
- Very hard expert questions across many subjectsHumanity's Last Exam+5.7points ahead
Where Claude Opus 4 pulls ahead
- Fixes real GitHub issues in Python projectsSWE-bench Verified+9.3points ahead
- Common-sense trick questionsSimpleBench+7.2points ahead
- Follows unfamiliar, precisely checkable instructionsIFBench+5points ahead
Every test, side by side
All 47 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingClaude Opus 4
- SWE-bench VerifiedClaude Opus 4 by 9.372.563.2+9.3
- SciCodeGemini 2.5 Pro by 6.539.846.3+6.5
- Terminal-Bench HardClaude Opus 4 by 4.631.126.5+4.6
ReasoningEven
- SimpleBenchClaude Opus 4 by 7.258.851.6+7.2
- Humanity's Last ExamGemini 2.5 Pro by 5.712.318+5.7
- GPQA DiamondGemini 2.5 Pro by 479.683.6+4
- ARC-AGI-2Claude Opus 4 by 3.78.64.9+3.7
FactsGemini 2.5 Pro
- Vectara HHEM hallucination ratelower is betterGemini 2.5 Pro by 5127+5
Images and chartsGemini 2.5 Pro
- LMArena · VisionGemini 2.5 Pro by 71 rating points11921263+71 rating
- MMMUGemini 2.5 Pro by 5.973.779.6+5.9
Long documentsGemini 2.5 Pro
- LongBench v2Gemini 2.5 Pro by 3.255.658.8+3.2
- AA-LCRtie69.369tie
Following instructionsClaude Opus 4
- IFBenchClaude Opus 4 by 553.748.7+5
- Multi-ChallengeClaude Opus 4 by 558.653.6+5
Other results33 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 2025Gemini 2.5 Pro by 49.815.965.7+49.8
- AIME 2024Gemini 2.5 Pro by 43.848.292+43.8
- ZebraLogicGemini 2.5 Pro by 32.359.391.6+32.3
- SimpleQAGemini 2.5 Pro by 2822.850.8+28
- OpenAI-MRCR (128k)Gemini 2.5 Pro by 27.948.976.8+27.9
- HarmBenchClaude Opus 4 by 26.391.765.4+26.3
- livecodebench_hardGemini 2.5 Pro by 26.333.159.4+26.3
- HMMT Feb. 2025Gemini 2.5 Pro by 20.86080.8+20.8
- LiveCodeBench (24/8~25/5)Gemini 2.5 Pro by 20.556.677.1+20.5
- τ²-Bench Telecom (AA run)Claude Opus 4 by 19.373.454.1+19.3
- Terminal-BenchClaude Opus 4 by 17.943.225.3+17.9
- TAU-bench (retail)Claude Opus 4 by 14.481.467+14.4
- Artificial Analysis Coding IndexGemini 2.5 Pro by 12.73446.7+12.7
- AIR-Bench 2024Claude Opus 4 by 12.185.773.6+12.1
- livecodebench_mediumGemini 2.5 Pro by 10.280.490.6+10.2
- TAU-bench (airline)Claude Opus 4 by 9.659.650+9.6
- vectara_answer_rateGemini 2.5 Pro by 8.19199.1+8.1
- AIME 2025Gemini 2.5 Pro by 7.575.583+7.5
- Aider-PolyglotGemini 2.5 Pro by 5.870.776.5+5.8
- AA IntelligenceClaude Opus 4 by 5.620.615+5.6
- vectara_factual_consistencyGemini 2.5 Pro by 58893+5
- MATH-500 (EM)Gemini 2.5 Pro by 4.494.498.8+4.4
- LiveCodeBenchGemini 2.5 Pro by 3.870.474.2+3.8
- simple_safety_testsClaude Opus 4 by 310097+3
- τ²-BenchClaude Opus 4 by 35754+3
- XSTestGemini 2.5 Pro by 1.79798.7+1.7
- vectara_avg_summary_lengthClaude Opus 4 by 16.8 rating points123.2106.4+16.8 rating
- ARC-AGI-1Gemini 2.5 Pro by 1.335.737+1.3
- bbqtie97.196.4tie
- anthropic_red_teamtie98.999.5tie
- MMLU-Protie86.686tie
- livecodebench_easytie99.198.8tie
- frontiermath_tier_4_v1tie4.24.2tie
Questions people ask
Which is better, Claude Opus 4 or Gemini 2.5 Pro?
Gemini 2.5 Pro wins three of the six areas where both have results: facts, images and charts and long documents. Claude Opus 4 wins coding and following instructions. They are level on reasoning.
Which is better for coding?
Claude Opus 4. It wins 2 of the 3 coding tests both models report; Gemini 2.5 Pro wins 1.
Which is cheaper?
Claude Opus 4 costs $15.00 per million input tokens and $75.00 per million output tokens; Gemini 2.5 Pro costs $1.25 and $10.00. That makes Gemini 2.5 Pro about 88% cheaper for the same work.
How do you compare the two?
We use the 47 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 33 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.