Estimate what a workload costs on the Claude API — at Anthropic’s official list prices and at our gateway rates — then see the difference. All prices are per million tokens.
| Model | Context | Official in / out | Our in / out | Official / mo | Ours / mo | You save |
|---|---|---|---|---|---|---|
| Claude Opus 4.8 | 200K tokens | $15 / $75 | $3 / $15 | — | — | — |
| Claude Fable 5 | 200K · 1M variant | $15 / $75 | $3 / $15 | — | — | — |
| GPT-5.5 | 256K tokens | $10 / $40 | $3 / $12 | — | — | — |
| Claude 1M Context | 1,000,000 tokens | $6 / $22.5 | $1.8 / $6.75 | — | — | — |
| GPT-5 | 256K tokens | $5 / $20 | $1.5 / $6 | — | — | — |
| Claude Sonnet 4.6 | 200K · 1M beta | $3 / $15 | $0.9 / $4.5 | — | — | — |
| Gemini 3 Pro | 1M tokens | $2.5 / $15 | $1 / $6 | — | — | — |
| MiniMax M3 | 1M tokens | $1.2 / $6 | $0.5 / $2.5 | — | — | — |
| Claude Haiku 4.5 | 200K tokens | $1 / $5 | $0.4 / $2 | — | — | — |
| Gemini 3 Flash | 1M tokens | $0.5 / $3 | $0.2 / $1.2 | — | — | — |
Official and gateway rates are both per million tokens. Monthly figures multiply your inputs by the per-token rates; caching assumes 80% of input tokens are cache hits.
Claude is billed per million tokens, priced separately for input and output. At official list prices the flagship Opus tier is $15 input and $75 output per million tokens; mid-tier and small models cost substantially less. Our gateway prices the same models at $3 and $15.
It depends entirely on volume. A subscription is a fixed monthly fee for chat use, while the API bills per token and can cost far less for light use or far more for heavy automated use. Use the calculator above with your real call volume, and if the API number is consistently high, a flat-rate unlimited plan is usually cheaper than both.
Route easy tasks to Haiku instead of Opus, use the Batch API for anything that can wait up to 24 hours, cache long system prompts and shared context, trim retrieved context before sending it, and cap max_tokens so runaway outputs cannot bill unexpectedly.
Yes — the same model IDs, the same context windows and the same request and response formats. The gateway is a drop-in base-URL change; your existing Anthropic or OpenAI SDK code keeps working unmodified.
No. The table compares per-token rates only, so it is a like-for-like comparison against Anthropic's list prices. The 10x top-up multiplier applies on top of that when you buy credit, so real spend is lower again.
No. Pay-as-you-go credit stays on the balance until it is consumed. Flat-rate unlimited plans are different — those run for a fixed period and simply end, with no auto-renewal.
Same models, same context windows, one drop-in base URL. Pay as you go or take a flat-rate unlimited plan.
Get an API key See unlimited plansClaudeAPIKey.dev is an independent API gateway and is not affiliated with, endorsed by, or sponsored by Anthropic. “Claude” and “Anthropic” are trademarks of Anthropic. Official Anthropic list prices shown for comparison are taken from Anthropic’s published pricing and may change; our rates are read from our own live catalogue. Estimates are calculated from the token counts you enter and will differ from real usage.