Claude 3 Haiku wins more areas, narrowly.GPT-4 is better at facts.
Scores updated · 18 tests both models report · How we compare
Where each one wins
Tests won in each of the four 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.
Agents, reasoning and following instructions 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 3 Haiku costs 98% 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 Claude 3 Haiku pulls ahead
- Follows unfamiliar, precisely checkable instructionsIFBench+2.9points ahead
- Graduate-level biology, physics and chemistry questionsGPQA Diamond+2.5points ahead
Where GPT-4 pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+11.1points ahead
- Answers hard knowledge questions correctlyAA-Omniscience · Accuracy+3.4points ahead
Every test, side by side
All 18 tests both models report. The winning score is in its model's colour; marks a score checked independently.
FactsGPT-4
- AA-Omniscience · Non-hallucinationGPT-4 by 11.119.530.6+11.1
- AA-Omniscience · AccuracyGPT-4 by 3.417.621+3.4
Following instructionsClaude 3 Haiku
- IFBenchClaude 3 Haiku by 2.936.133.2+2.9
Other results13 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.
- NarrativeQAGPT-4 by 52.424.476.8+52.4
- naturalquestions_closedbookGPT-4 by 31.314.445.7+31.3
- AIR-Bench 2024Claude 3 Haiku by 18.582.764.2+18.5
- AA-OmniscienceGPT-4 by 14.8-48.6-33.8+14.8
- OpenBookQAGPT-4 by 12.283.896+12.2
- HellaSwagGPT-4 by 9.485.995.3+9.4
- HumanEvalClaude 3 Haiku by 8.975.967+8.9
- MMLUGPT-4 by 7.366.273.5+7.3
- Artificial Analysis Coding IndexGPT-4 by 6.46.713.1+6.4
- GSM8KGPT-4 by 4.388.993.2+4.3
- DROPGPT-4 by 2.578.480.9+2.5
- AA IntelligenceGPT-4 by 1.15.66.7+1.1
- MGSMtie75.174.5tie
Questions people ask
Which is better, Claude 3 Haiku or GPT-4?
Claude 3 Haiku wins two of the four areas where both have results: reasoning and following instructions. GPT-4 wins facts. They are level on agents.
Which is cheaper?
Claude 3 Haiku costs $0.25 per million input tokens and $1.25 per million output tokens; GPT-4 costs $30.00 and $60.00. That makes Claude 3 Haiku about 98% cheaper for the same work.
How do you compare the two?
We use the 18 benchmark tests both models have published scores on. The verdict counts the 5 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 13 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.