Compare AI models

Specifications
Context window
1M400k
API pricing
Input price
$0.10$1.75
Output price
$0.50$14.00
Cached input price
$0.01$0.175
Cheapest input
$0.10Amazon Bedrock$1.75Azure
Cheapest output
$0.50Amazon Bedrock$14.00Azure
$45.00Anthropic list price$1,050OpenAI list price
Benchmarks
BullshitBench v2
57%24%
SWE-Bench Pro
64.8%56.8%
threejseval
14241153
Overview
CompanyAnthropicOpenAI
Release dateOct 7 2026Feb 5 2026
AccessClosedClosed
Model detailsView modelView model

Frequently asked questions

Claude Haiku 5.5 leads GPT-5.3-Codex on 3 of the 3 benchmarks they both report (BullshitBench v2, SWE-Bench Pro, threejseval). Claude Haiku 5.5 is cheaper on both input and output: $0.10 vs $1.75 per million input tokens, and $0.50 vs $14.00 per million output tokens. Figures are base-tier rates. GPT-5.3-Codex shipped 244 days before Claude Haiku 5.5, so benchmark comparisons should account for the intervening progress.

Context windows are 1M (Claude Haiku 5.5) vs 400k (GPT-5.3-Codex).

On BullshitBench v2, Claude Haiku 5.5 leads at 57% vs GPT-5.3-Codex at 24%. On SWE-Bench Pro, Claude Haiku 5.5 leads at 64.8% vs GPT-5.3-Codex at 56.8%. On threejseval, Claude Haiku 5.5 leads at 1424 vs GPT-5.3-Codex at 1153.