Compare AI models

Specifications
Context window
200k1.05M
API pricing
Input price
$1.00$8.00
Output price
$5.00$40.00
Cached input price
$0.10$0.40
Cheapest input
$1.00Amazon Bedrock—
Cheapest output
$5.00Amazon Bedrock—
$450Anthropic list price$3,600OpenAI list price
Benchmarks
Terminal-Bench-Science 0.1
0.1%77.1%
GPQA Diamond
73%97%
Overview
CompanyAnthropicOpenAI
Release dateOct 15 2025Nov 17 2026
AccessClosedClosed
Model detailsView modelView model

Frequently asked questions

GPT-6.2 Astra leads Claude Haiku 4.5 on 2 of the 2 benchmarks they both report (Terminal-Bench-Science 0.1, GPQA Diamond). Claude Haiku 4.5 is cheaper on both input and output: $1.00 vs $8.00 per million input tokens, and $5.00 vs $40.00 per million output tokens. Figures are base-tier rates. Claude Haiku 4.5 shipped 398 days before GPT-6.2 Astra, so benchmark comparisons should account for the intervening progress.

Context windows are 200k (Claude Haiku 4.5) vs 1.05M (GPT-6.2 Astra).

On Terminal-Bench-Science 0.1, GPT-6.2 Astra leads at 77.1% vs Claude Haiku 4.5 at 0.1%. On GPQA Diamond, GPT-6.2 Astra leads at 97% vs Claude Haiku 4.5 at 73%.