Pricing / Prompt caching
Cheapest price per million input tokens for Prompt caching
Unit: usd_per_million_input_tokens only — nothing converted or blended. min sourced input SKU among candidates; not quality-adjusted. Updated 2026-10-27 (newest asOf / re-check on this page).
Pick is not a price winner. anthropic sourced input $/1M is 0.1; cheapest among candidates is amazon-bedrock at 0.04.
| # | Provider | SKU | Input $/1M | Source | asOf / re-checked | First seen / last changed |
|---|---|---|---|---|---|---|
| 1 | Amazon Bedrock | Gemma 3 4B | $0.04 | page | 2026-10-03 / 2026-10-27 | 2026-10-03 / no change recorded |
| 2 | AnthropicDepot Pick | Haiku 5.5 (prompts up to 100K tokens) | $0.1 | page | 2026-10-27 / 2026-10-27 | 2026-10-21 / no change recorded |
| 3 | OpenAI | gpt-6-luna | $0.1 | page | 2026-10-03 / 2026-10-27 | 2026-10-03 / no change recorded |
| 4 | Google AI (Gemini) | gemini-3.5-flash-lite | $0.3 | page | 2026-10-21 / not reconfirmed | 2026-10-21 / no change recorded |
Price only: not quality-, latency- or limit-adjusted. The Depot Pick is editorial and is not chosen by price. Brief: /jtbds/prompt-caching-optimization · Agents: GET /api/quote?jtbd=prompt-caching-optimization