Comparison

Claude Sonnet 5 vs GPT-5.5

Per input token, Claude Sonnet 5 runs 60% cheaper than GPT-5.5. Claude Sonnet 5 is 67% cheaper on output tokens. GPT-5.5 has the larger context window than Claude Sonnet 5.

FieldClaude Sonnet 5
GPT-5.5
ProviderAnthropicOpenAI
Context window1M1.1M
Max output128K128K
Input / 1M$2.0060% cheaper$5.00
Output / 1M$10.0067% cheaper$30.00
Cached input / 1M$0.2060% cheaper$0.50
Batch discount50%50%
Release date2026-04-23

Worked cost examples

Estimated monthly cost on three common usage shapes

WorkloadClaude Sonnet 5GPT-5.5

Chatbot

800 input / 200 output tokens per request, 100,000 requests/month

$360.00cheaper$1,000

RAG pipeline

6,000 input / 700 output tokens per request, 30,000 requests/month

$570.00cheaper$1,530

Batch summarization

20,000 input / 1,500 output tokens per request, 5,000 requests/month

$275.00cheaper$725.00

Claude Sonnet 5 vs GPT-5.5 FAQ

Is Claude Sonnet 5 cheaper than GPT-5.5?
Claude Sonnet 5 costs $2.00 per 1M input tokens versus $5.00 for GPT-5.5 — Claude Sonnet 5 is 60% cheaper per input token.
Which is cheaper for output tokens, Claude Sonnet 5 or GPT-5.5?
Claude Sonnet 5 charges $10.00 per 1M output tokens; GPT-5.5 charges $30.00 per 1M output tokens.
Which has the larger context window, Claude Sonnet 5 or GPT-5.5?
GPT-5.5 does, with a 1,050,000-token context window.
Do Claude Sonnet 5 and GPT-5.5 offer cache pricing?
Claude Sonnet 5 discounts cached input to $0.20 per 1M tokens. GPT-5.5 discounts cached input to $0.50 per 1M tokens.

Looking for another matchup? See all comparisons.