AI & LLM
Token, cost & context tools for LLMs
Claude API Cost Calculator — Anthropic Claude Opus 4.8 Pricing
Claude Opus 4.8 (Anthropic's flagship model as of 2026-07-03) is priced at $5.00/M input tokens and $25.00/M output tokens, with prompt caching at $0.50/M for cached input. At the reference workload of 1,000 input + 500 output tokens per API call, the cost is $0.02: $0.005 input + $0.01 output. Output tokens cost 5× more than input. At 1,000 calls/day, that is $17.50/day. Use the full calculator with your actual token counts for a precise monthly estimate.
Reference workload: 1,000 input + 500 output tokens
- Cost per call: $0.02 (input $0.005 + output $0.01)
- Per 1,000 calls: $17.50
- Input: $5.00/M | Cached input: $0.50/M | Output: $25.00/M
- Output costs 5× more than input
- Rates as of 2026-07-03 (Claude Opus 4.8)
Frequently asked questions
How much does Claude Opus 4.8 cost per API call?
At 1000 input + 500 output tokens — a typical mid-length chat turn — Claude Opus 4.8 costs $0.02/call and $17.50 per 1,000 calls. Input is $5.00/M; output is $25.00/M. With prompt caching on repeated system prompts, the effective input cost drops to $0.50/M for the cached portion. Rates as of 2026-07-03.
How does Claude Opus 4.8 pricing compare to GPT-5.5?
Claude Opus 4.8 and GPT-5.5 have the same input price ($5.00/M) but Claude charges $25.00/M for output vs GPT-5.5's $30.00/M — making Claude cheaper for output-heavy workloads. Both offer prompt caching at a ~90% discount on cached input. Gemini 2.5 Pro is the lowest-cost option at $1.25/M input and $10.00/M output.