Claude Haiku 4.5 vs Gemini 2.5 Pro
Wins 3 of 7 areas
Coding · Agents · Long documents
Wins 3 of 7 areas
Reasoning · Facts · Images and charts
The two are evenly matched.Claude Haiku 4.5 is better at coding and agents; Gemini 2.5 Pro at reasoning and facts.
Scores updated · 38 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.
- ReasoningHard problems that need careful thinking03Gemini 2.5 Pro3 of 4 tests · 1 tie
- FactsGetting facts right instead of making them up13Gemini 2.5 Pro3 of 4 tests
- Images and chartsUnderstanding pictures, charts and video01Gemini 2.5 Pro1 of 1 test
- CodingWriting and fixing software31Claude Haiku 4.53 of 5 tests · 1 tie
- AgentsCarrying out multi-step tasks on its own10Claude Haiku 4.51 of 3 tests · 2 ties
- Long documentsFinding answers in very long texts10Claude Haiku 4.51 of 1 test
- Following instructionsDoing exactly what it is asked11Even1 each
Images and charts and long documents 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.
Claude Haiku 4.5 costs 47% 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
- Short factual questions, answered correctlySimpleQA Verified+42.8points ahead
- Answers hard knowledge questions correctlyAA-Omniscience · Accuracy+21.1points ahead
- Graduate-level biology, physics and chemistry questionsGPQA Diamond+16.4points ahead
Where Claude Haiku 4.5 pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+63.6points ahead
- Command-line tasks in a real terminalTerminal-Bench 2.1+15.7points ahead
- Fixes real GitHub issues in Python projectsSWE-bench Verified+10.1points ahead
Every test, side by side
All 38 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingClaude Haiku 4.5
- Terminal-Bench 2.1Claude Haiku 4.5 by 15.744.228.5+15.7
- LMArena · WebDevClaude Haiku 4.5 by 105 rating points13291224+105 rating
- SWE-bench VerifiedClaude Haiku 4.5 by 10.173.363.2+10.1
- SciCodeGemini 2.5 Pro by 4.142.246.3+4.1
- Terminal-Bench Hardtie27.326.5tie
AgentsClaude Haiku 4.5
- GDPValClaude Haiku 4.5 by 11.811.80+11.8
- τ-Bench V3 · Bankingtie9.39.7tie
- Terminal-Bench 4.0tie00tie
ReasoningGemini 2.5 Pro
- GPQA DiamondGemini 2.5 Pro by 16.467.283.6+16.4
- Humanity's Last ExamGemini 2.5 Pro by 7.610.418+7.6
- CritPtGemini 2.5 Pro by 2.602.6+2.6
- ARC-AGI-2tie44.9tie
FactsGemini 2.5 Pro
- AA-Omniscience · Non-hallucinationClaude Haiku 4.5 by 63.672.79.1+63.6
- SimpleQA VerifiedGemini 2.5 Pro by 42.813.256+42.8
- AA-Omniscience · AccuracyGemini 2.5 Pro by 21.11839+21.1
- Vectara HHEM hallucination ratelower is betterGemini 2.5 Pro by 2.89.87+2.8
Images and chartsGemini 2.5 Pro
- MMMU-ProGemini 2.5 Pro by 16.358.674.9+16.3
Following instructionsEven
- IFBenchClaude Haiku 4.5 by 5.654.348.7+5.6
- Multi-ChallengeGemini 2.5 Pro by 3.150.553.6+3.1
Other results18 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.
- HarmBenchClaude Haiku 4.5 by 30.595.965.4+30.5
- AIR-Bench 2024Claude Haiku 4.5 by 19.693.273.6+19.6
- Terminal-BenchClaude Haiku 4.5 by 15.74125.3+15.7
- AA-OmniscienceClaude Haiku 4.5 by 11.9-4.4-16.4+11.9
- ARC-AGI-1Claude Haiku 4.5 by 10.747.737+10.7
- AA Agentic IndexClaude Haiku 4.5 by 6.810.33.5+6.8
- XSTestGemini 2.5 Pro by 5.593.298.7+5.5
- bbqGemini 2.5 Pro by 3.692.896.4+3.6
- Artificial Analysis Coding IndexGemini 2.5 Pro by 2.843.946.7+2.8
- vectara_factual_consistencyGemini 2.5 Pro by 2.890.293+2.8
- anthropic_red_teamGemini 2.5 Pro by 2.696.999.5+2.6
- AIME 2025Gemini 2.5 Pro by 2.380.783+2.3
- frontiermath_tier_4_v1Gemini 2.5 Pro by 2.12.14.2+2.1
- AA IntelligenceClaude Haiku 4.5 by 1.916.915+1.9
- simple_safety_testsClaude Haiku 4.5 by 1.898.897+1.8
- vectara_avg_summary_lengthtie115.1106.4tie
- τ²-Bench Telecom (AA run)tie54.754.1tie
- vectara_answer_ratetie99.599.1tie
Questions people ask
Which is better, Claude Haiku 4.5 or Gemini 2.5 Pro?
Claude Haiku 4.5 and Gemini 2.5 Pro each win three of the seven areas where both have results. Claude Haiku 4.5 wins coding, agents and long documents; Gemini 2.5 Pro wins reasoning, facts and images and charts. Claude Haiku 4.5 costs 47% less. They are level on following instructions.
Which is better for coding?
Claude Haiku 4.5. It wins 3 of the 5 coding tests both models report; Gemini 2.5 Pro wins 1, and 1 is a tie.
Which is cheaper?
Claude Haiku 4.5 costs $1.00 per million input tokens and $5.00 per million output tokens; Gemini 2.5 Pro costs $1.25 and $10.00. That makes Claude Haiku 4.5 about 47% cheaper for the same work.
How do you compare the two?
We use the 38 benchmark tests both models have published scores on. The verdict counts the 20 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 18 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.