Changelog
Every pricing change, model add, and deprecation we track. Pricing accuracy is the product; this page is how we keep ourselves honest.
Data confidence
"Inferred" entries are best-effort approximations, typically within 10% of true rates but may be wrong. Use the report incorrect price link if you spot an error.
Recent changes
- pricing
Anthropic verification pass against platform.claude.com/docs: Sonnet 4.6 ($3/$15) and Haiku 4.5 ($1/$5) confirmed at current rates. Refreshed Opus 4.7 to the current Opus 4.5+ tier ($5/$25 per 1M, input/output); cache write/read, batch, and 1M-context fields updated to match. All three Anthropic rows now flagged "official". OpenAI, Google, and Meta verification passes pending.
- plans
Verified Claude Code plan prices against claude.com/pricing: Pro $20/mo (or $17/mo billed annually at $200), Max from $100/mo, Max 20x at $200/mo. Plan data flipped to "official".
- launch
Initial pricing data seeded for 10 models across Anthropic, OpenAI, Google, and Meta. All entries marked as inferred. Verification pass against official provider docs is the next priority before any production use of these numbers.
Tracked models
| Model | Confidence | Last verified |
|---|---|---|
| Claude Opus 4.7 | official | 2026-05-06 |
| Claude Sonnet 4.6 | official | 2026-05-06 |
| Claude Haiku 4.5 | official | 2026-05-06 |
| GPT-4o | inferred | 2026-04-28 |
| GPT-4o mini | inferred | 2026-04-28 |
| OpenAI o3 | inferred | 2026-04-28 |
| OpenAI o3-mini | inferred | 2026-04-28 |
| Gemini 2.5 Pro | inferred | 2026-04-28 |
| Gemini 2.5 Flash | inferred | 2026-04-28 |
| Llama 3.3 70B | inferred | 2026-04-28 |