LLM prices / Compare
API price comparison · checked Oct 23, 2026
Claude Sonnet 5.5 vs Claude Opus 4.8: API price
Short answerClaude Sonnet 5.5 is the cheaper one: 1,000 chatbot replies cost $7.00 against $17.50 on Claude Opus 4.8, so Claude Opus 4.8 costs 2.50x as much. Across our 6 workloads Claude Sonnet 5.5 is cheaper on 6 and Claude Opus 4.8 on 0. Standard prices: Claude Sonnet 5.5 $2 / $10, Claude Opus 4.8 $5 / $25 per 1M input / output tokens, checked Oct 23, 2026.
Cost of the same work
| Workload (1,000 requests) | Claude Sonnet 5.5 | Claude Opus 4.8 | Ratio |
|---|---|---|---|
| Chatbot replya 1,500-token prompt (1,000 of it a fixed system prompt) and a 400-token answer | $7.00 | $17.50 | 0.40x |
| Answer from your docs (RAG)6,000 tokens of retrieved context plus the question, a 500-token answer | $17.00 | $42.50 | 0.40x |
| Summarize a long documenta 20,000-token document in, an 800-token summary out | $48.00 | $120 | 0.40x |
| Coding agent step40,000 tokens of repo context and history (30,000 of it repeated from the last step), 2,000 out | $100 | $250 | 0.40x |
| Batch extraction2,000 tokens per record in, 300 tokens of JSON out, sent through the batch API where there is one | $3.50 | $8.75 | 0.40x |
| Long-document Q&Aa 150,000-token contract or codebase in one prompt, a 1,000-token answer | $310 | $775 | 0.40x |
Side by side
| Price | Claude Sonnet 5.5 | Claude Opus 4.8 |
|---|---|---|
| Input / 1M | $2 | $5 |
| Output / 1M | $10 | $25 |
| Cache read / 1M | $0.1 | $0.5 |
| Batch input / output | $1 / $5 | $2.5 / $12.5 |
| Long-prompt tier | none listed | none listed |
| Provider | Anthropic | Anthropic |
| Status | listed | listed |
When each one is cheaper
- Claude Sonnet 5.5 is cheaper or equal on both input and output, so it costs less on every workload at standard prices.
- With caching on a coding-agent step (30,000 of 40,000 input tokens cached): Claude Sonnet 5.5 $43.00 vs Claude Opus 4.8 $115 per 1,000 steps (0.37x).
- This page compares price only. Which model answers better for your task is a separate question: test both on your own prompts.
Each model in full
Sources
- Anthropic API pricing: https://platform.claude.com/docs/en/about-claude/pricing, checked Oct 23, 2026.