Claude Haiku 4.5 vs Gemini 3 Pro
Wins 0 of 7 areas
—
Wins 6 of 7 areas
Coding · Agents · Reasoning · Images and charts · Long documents · Following instructions
Gemini 3 Pro is the stronger all-rounder.Claude Haiku 4.5 is cheaper.
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.
- CodingWriting and fixing software15Gemini 3 Pro5 of 6 tests
- ReasoningHard problems that need careful thinking04Gemini 3 Pro4 of 4 tests
- Following instructionsDoing exactly what it is asked02Gemini 3 Pro2 of 2 tests
- AgentsCarrying out multi-step tasks on its own01Gemini 3 Pro1 of 1 test
- Images and chartsUnderstanding pictures, charts and video01Gemini 3 Pro1 of 1 test
- Long documentsFinding answers in very long texts01Gemini 3 Pro1 of 1 test
- FactsGetting facts right instead of making them up22Even2 each
Agents, 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 57% 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 3 Pro pulls ahead
- Short factual questions, answered correctlySimpleQA Verified+59.7points ahead
- Answers hard knowledge questions correctlyAA-Omniscience · Accuracy+37.8points ahead
- Very hard expert questions across many subjectsHumanity's Last Exam+29.3points ahead
Where Claude Haiku 4.5 pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+64.2points ahead
- Fixes real GitHub issues in many programming languagesSWE-bench Multilingual+2.4points 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.
CodingGemini 3 Pro
- Terminal-Bench HardGemini 3 Pro by 14.427.341.7+14.4
- SciCodeGemini 3 Pro by 13.942.256.1+13.9
- LMArena · WebDevGemini 3 Pro by 110 rating points13291439+110 rating
- SWE-bench ProGemini 3 Pro by 3.839.543.3+3.8
- SWE-bench VerifiedGemini 3 Pro by 2.973.376.2+2.9
- SWE-bench MultilingualClaude Haiku 4.5 by 2.467.465+2.4
ReasoningGemini 3 Pro
- Humanity's Last ExamGemini 3 Pro by 29.310.439.7+29.3
- ARC-AGI-2Gemini 3 Pro by 27.1431.1+27.1
- GPQA DiamondGemini 3 Pro by 23.667.290.8+23.6
- CritPtGemini 3 Pro by 9.109.1+9.1
FactsEven
- AA-Omniscience · Non-hallucinationClaude Haiku 4.5 by 64.272.78.5+64.2
- SimpleQA VerifiedGemini 3 Pro by 59.713.272.9+59.7
- AA-Omniscience · AccuracyGemini 3 Pro by 37.81855.8+37.8
- Vectara HHEM hallucination ratelower is betterClaude Haiku 4.5 by 3.89.813.6+3.8
Images and chartsGemini 3 Pro
- MMMU-ProGemini 3 Pro by 21.658.680.2+21.6
Following instructionsGemini 3 Pro
- IFBenchGemini 3 Pro by 16.154.370.4+16.1
- Multi-ChallengeGemini 3 Pro by 15.250.565.7+15.2
Other results19 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.
- AA Agentic IndexGemini 3 Pro by 41.710.352+41.7
- τ²-Bench Telecom (AA run)Gemini 3 Pro by 32.454.787.1+32.4
- ARC-AGI-1Gemini 3 Pro by 27.347.775+27.3
- HarmBenchClaude Haiku 4.5 by 23.495.972.5+23.4
- AIR-Bench 2024Claude Haiku 4.5 by 2093.273.2+20
- AA-OmniscienceGemini 3 Pro by 19.7-4.415.3+19.7
- AIME 2025Gemini 3 Pro by 19.380.7100+19.3
- frontiermath_tier_4_v1Gemini 3 Pro by 16.72.118.8+16.7
- AA IntelligenceGemini 3 Pro by 11.116.928+11.1
- MMMLUGemini 3 Pro by 8.88391.8+8.8
- bbqGemini 3 Pro by 5.692.898.4+5.6
- XSTestGemini 3 Pro by 4.193.297.3+4.1
- vectara_factual_consistencyClaude Haiku 4.5 by 3.890.286.4+3.8
- Artificial Analysis Coding IndexGemini 3 Pro by 2.643.946.5+2.6
- τ²-Bench (Retail)Gemini 3 Pro by 2.183.285.3+2.1
- vectara_avg_summary_lengthClaude Haiku 4.5 by 13.2 rating points115.1101.9+13.2 rating
- simple_safety_testsClaude Haiku 4.5 by 1.398.897.5+1.3
- anthropic_red_teamtie96.997tie
- vectara_answer_ratetie99.599.4tie
Questions people ask
Which is better, Claude Haiku 4.5 or Gemini 3 Pro?
Gemini 3 Pro wins six of the seven areas where both have results: coding, agents, reasoning, images and charts, long documents and following instructions. Claude Haiku 4.5 wins none, but costs 57% less. They are level on facts.
Which is better for coding?
Gemini 3 Pro. It wins 5 of the 6 coding tests both models report; Claude Haiku 4.5 wins 1.
Which is cheaper?
Claude Haiku 4.5 costs $1.00 per million input tokens and $5.00 per million output tokens; Gemini 3 Pro costs $2.00 and $12.00. That makes Claude Haiku 4.5 about 57% 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 19 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 19 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.