Claude Haiku 4.5 vs GPT-5
Wins 0 of 7 areas
—
Wins 6 of 7 areas
Coding · Agents · Reasoning · Images and charts · Long documents · Following instructions
GPT-5 is the stronger all-rounder.Claude Haiku 4.5 is cheaper.
Scores updated · 42 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 thinking04GPT-54 of 4 tests
- CodingWriting and fixing software24GPT-54 of 7 tests · 1 tie
- AgentsCarrying out multi-step tasks on its own02GPT-52 of 2 tests
- Following instructionsDoing exactly what it is asked02GPT-52 of 2 tests
- Images and chartsUnderstanding pictures, charts and video01GPT-51 of 1 test
- Long documentsFinding answers in very long texts01GPT-51 of 1 test
- FactsGetting facts right instead of making them up22Even2 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 GPT-5 pulls ahead
- Short factual questions, answered correctlySimpleQA Verified+36.9points ahead
- Answers hard knowledge questions correctlyAA-Omniscience · Accuracy+22.3points ahead
- Keeps track of context across a multi-turn chatMulti-Challenge+19.1points ahead
Where Claude Haiku 4.5 pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+54.9points ahead
- Fixes real GitHub issues in many programming languagesSWE-bench Multilingual+12.1points ahead
- Command-line tasks in a real terminalTerminal-Bench 2.1+9points ahead
Every test, side by side
All 42 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingGPT-5
- SWE-bench MultilingualClaude Haiku 4.5 by 12.167.455.3+12.1
- Terminal-Bench 2.1Claude Haiku 4.5 by 944.235.2+9
- LMArena · WebDevGPT-5 by 88 rating points13291417+88 rating
- Terminal-Bench HardGPT-5 by 5.327.332.6+5.3
- SWE-bench ProGPT-5 by 2.339.541.8+2.3
- SWE-bench VerifiedGPT-5 by 1.673.374.9+1.6
- SciCodetie42.242.9tie
AgentsGPT-5
- τ-Bench V3 · BankingGPT-5 by 12.89.322.1+12.8
- GDPValGPT-5 by 9.211.821+9.2
ReasoningGPT-5
- GPQA DiamondGPT-5 by 18.267.285.4+18.2
- Humanity's Last ExamGPT-5 by 18.110.428.5+18.1
- ARC-AGI-2GPT-5 by 5.949.9+5.9
- CritPtGPT-5 by 5.705.7+5.7
FactsEven
- AA-Omniscience · Non-hallucinationClaude Haiku 4.5 by 54.972.717.8+54.9
- SimpleQA VerifiedGPT-5 by 36.913.250.1+36.9
- AA-Omniscience · AccuracyGPT-5 by 22.31840.3+22.3
- Vectara HHEM hallucination ratelower is betterClaude Haiku 4.5 by 5.39.815.1+5.3
Following instructionsGPT-5
- Multi-ChallengeGPT-5 by 19.150.569.6+19.1
- IFBenchGPT-5 by 18.854.373.1+18.8
Other results21 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)GPT-5 by 30.154.784.8+30.1
- ARC-AGI-1GPT-5 by 1847.765.7+18
- AA Agentic IndexGPT-5 by 16.210.326.5+16.2
- HealthBench Professional length-adjustedGPT-5 by 1432.246.2+14
- AIME 2025GPT-5 by 13.980.794.6+13.9
- frontiermath_tier_4_v1GPT-5 by 10.42.112.5+10.4
- AA IntelligenceGPT-5 by 6.116.923+6.1
- Artificial Analysis Coding IndexClaude Haiku 4.5 by 6.143.937.8+6.1
- AIR-Bench 2024Claude Haiku 4.5 by 5.593.287.7+5.5
- vectara_factual_consistencyClaude Haiku 4.5 by 5.390.284.9+5.3
- vectara_avg_summary_lengthGPT-5 by 47.6 rating points115.1162.7+47.6 rating
- AA-OmniscienceClaude Haiku 4.5 by 4.3-4.4-8.7+4.3
- bbqGPT-5 by 492.896.8+4
- XSTestGPT-5 by 3.993.297.1+3.9
- Terminal-BenchGPT-5 by 2.84143.8+2.8
- anthropic_red_teamGPT-5 by 2.296.999.1+2.2
- τ²-Bench (Retail)Claude Haiku 4.5 by 2.183.281.1+2.1
- HarmBenchGPT-5 by 1.795.997.6+1.7
- simple_safety_testsGPT-5 by 198.899.8+1
- Tau2 airlineClaude Haiku 4.5 by 163.662.6+1
- vectara_answer_ratetie99.599.9tie
Questions people ask
Which is better, Claude Haiku 4.5 or GPT-5?
GPT-5 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 47% less. They are level on facts.
Which is better for coding?
GPT-5. It wins 4 of the 7 coding tests both models report; Claude Haiku 4.5 wins 2, 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; GPT-5 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 42 benchmark tests both models have published scores on. The verdict counts the 21 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 21 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.