Input
$1.00
Output
$5.00
Source: Official pricing
Last checked: 2026-03-20
Claude Haiku 4.5 vs GPT-5.2 Comparison & Evaluation
Direct head-to-head results for this model pair.
Compare Performance by Model
This page summarizes direct comparisons between two models across standard tasks and discussions.
Overall (Tasks + Discussions)
Win Rate 25%
Wins 4
Draws 0
Losses 12
Standard Task Comparison
Win Rate 27%
Wins 3
Draws 0
Losses 8
Discussion Comparison
Win Rate 20%
Wins 1
Draws 0
Losses 4
Overall (Tasks + Discussions)
Win Rate 75%
Wins 12
Draws 0
Losses 4
Standard Task Comparison
Win Rate 73%
Wins 8
Draws 0
Losses 3
Discussion Comparison
Win Rate 80%
Wins 4
Draws 0
Losses 1
Official Pricing Comparison
This section places the official pricing of both models side by side using standard text rates. Actual total cost can still change with output length and billing conditions, so this is best read as a quick comparison of baseline list pricing.
Input
$1.75
Output
$14.00
Source: Official pricing
Last checked: 2026-03-20
If you want a fuller view including measured cost and overall value, see the AI Pricing Comparison & Best Value Ranking.
AI Pricing ComparisonCriteria Breakdown
Standard
Appropriateness
A Claude Haiku 4.5
B GPT-5.2
Audience Fit
A Claude Haiku 4.5
B GPT-5.2
Clarity
A Claude Haiku 4.5
B GPT-5.2
Code Quality
A Claude Haiku 4.5
B GPT-5.2
Coherence
A Claude Haiku 4.5
B GPT-5.2
Completeness
A Claude Haiku 4.5
B GPT-5.2
Compression
A Claude Haiku 4.5
B GPT-5.2
Correctness
A Claude Haiku 4.5
B GPT-5.2
Coverage
A Claude Haiku 4.5
B GPT-5.2
Creativity
A Claude Haiku 4.5
B GPT-5.2
Diversity
A Claude Haiku 4.5
B GPT-5.2
Emotional Impact
A Claude Haiku 4.5
B GPT-5.2
Empathy
A Claude Haiku 4.5
B GPT-5.2
Ethics & Safety
A Claude Haiku 4.5
B GPT-5.2
Faithfulness
A Claude Haiku 4.5
B GPT-5.2
Helpfulness
A Claude Haiku 4.5
B GPT-5.2
Humor Effectiveness
A Claude Haiku 4.5
B GPT-5.2
Instruction Following
A Claude Haiku 4.5
B GPT-5.2
Logic
A Claude Haiku 4.5
B GPT-5.2
Naturalness
A Claude Haiku 4.5
B GPT-5.2
Originality
A Claude Haiku 4.5
B GPT-5.2
Persona Consistency
A Claude Haiku 4.5
B GPT-5.2
Persuasiveness
A Claude Haiku 4.5
B GPT-5.2
Practical Value
A Claude Haiku 4.5
B GPT-5.2
Quantity
A Claude Haiku 4.5
B GPT-5.2
Safety
A Claude Haiku 4.5
B GPT-5.2
Structure
A Claude Haiku 4.5
B GPT-5.2
Style Quality
A Claude Haiku 4.5
B GPT-5.2
Usefulness
A Claude Haiku 4.5
B GPT-5.2
Discussion
Clarity
A Claude Haiku 4.5
B GPT-5.2
Instruction Following
A Claude Haiku 4.5
B GPT-5.2
Logic
A Claude Haiku 4.5
B GPT-5.2
Persuasiveness
A Claude Haiku 4.5
B GPT-5.2
Rebuttal Quality
A Claude Haiku 4.5
B GPT-5.2
Matchups With Significant Performance Gaps
Tasks
The Android's First Lie
Type: Tasks / Winner: GPT-5.2
Tasks
Empathetic Response to a Struggling New Employee
Type: Tasks / Winner: GPT-5.2
Tasks
Explain Public-Key Cryptography to a Project Manager
Type: Tasks / Winner: GPT-5.2
Tasks
Historical Figures as Modern Roommates
Type: Tasks / Winner: GPT-5.2
Tasks
Reinventing the Public Library for the Digital Age
Type: Tasks / Winner: GPT-5.2
Tasks
Persuasive Email for a Four-Day Work Week Pilot
Type: Tasks / Winner: GPT-5.2
Fairness / How This Comparison Was Built
This page aggregates completed direct head-to-head comparisons for this model pair only. Judging follows the same fairness policy used across Orivel, and translated text is for display.
See fairness policy