Anthropic API pricing
Anthropic sells API access to the Claude model family. Prices are per 1M tokens, with prompt-cache reads at 10% of the input rate, cache writes at a premium, and a Batch API at half price.
Official price list: platform.claude.com
Current lineup
Claude Fable 5
Anthropic's flagship Claude 5 model — the Mythos-class tier above Opus.
$10.00 in / $50.00 out per 1M
Claude Mythos 5
The variant of Anthropic's top Claude 5 model available to approved organizations; priced identically to Fable 5.
$10.00 in / $50.00 out per 1M
Claude Opus 5
Anthropic's current Opus-line model.
$5.00 in / $25.00 out per 1M
Claude Sonnet 5
Anthropic's mid-size Claude 5 model — the price/performance default.
$2.00 in / $10.00 out per 1M
Claude Haiku 4.5
Anthropic's small, fast, low-cost model.
$1.00 in / $5.00 out per 1M
Claude Opus 4.8
The last Opus 4.x release before Opus 5.
$5.00 in / $25.00 out per 1M
Claude Sonnet 4.6
The last Sonnet 4.x release before Sonnet 5.
$3.00 in / $15.00 out per 1M
Claude Opus 4.7
Opus-line release of April 2026.
$5.00 in / $25.00 out per 1M
Claude Opus 4.6
Opus-line release of February 2026 — the first 1M-context Opus.
$5.00 in / $25.00 out per 1M
Claude Opus 4.5
The late-2025 release that cut Opus pricing 3x to $5/$25.
$5.00 in / $25.00 out per 1M
Claude Sonnet 4.5
The 2025 Sonnet workhorse; huge installed base.
$3.00 in / $15.00 out per 1M
Claude Opus 4.1
2025 Opus release at the original $15/$75 Opus pricing — retired from the Claude API; still sold via Bedrock and Google Cloud.
$15.00 in / $75.00 out per 1M
How Anthropic billing works
Prompt caching: cache reads bill at 0.1x the input rate; cache writes bill at 1.25x (5-minute TTL) or 2x (1-hour TTL) the input rate. For prompts reused even a handful of times, caching is the single biggest lever on a Claude bill.
Batch API: flat 50% off input and output, and it stacks with cache pricing.
Claude 4.6-and-later models (including the Claude 5 family) include a 1M-token context window at standard per-token rates — no long-context surcharge. Earlier models top out at 200K.
Two premium knobs: requests pinned to US inference (inference_geo: "us") bill a 1.1x multiplier on 4.6+ models, and the Fast-mode research preview (Opus 5 / Opus 4.8) bills 2x standard. Web search costs $10 per 1,000 searches on top of tokens.
Every Anthropic model we price
The full catalog — including dated snapshots and variants — exactly as Marginal prices incoming usage events.
| Cached input / 1M | ||||
|---|---|---|---|---|
| claude-3-7-sonnet-20250219 | $3.00 | $15.00 | 200K | $0.30 |
| claude-3-haiku-20240307 | $0.25 | $1.25 | 200K | $0.03 |
| claude-3-opus-20240229 | $15.00 | $75.00 | 200K | $1.50 |
| claude-4-opus-20250514 | $15.00 | $75.00 | 200K | $1.50 |
| claude-4-sonnet-20250514 | $3.00 | $15.00 | 1M | $0.30 |
| claude-fable-5 | $10.00 | $50.00 | 1M | $1.00 |
| claude-haiku-4-5 | $1.00 | $5.00 | 200K | $0.10 |
| claude-haiku-4-5-20251001 | $1.00 | $5.00 | 200K | $0.10 |
| claude-mythos-5 | $10.00 | $50.00 | 1M | $1.00 |
| claude-mythos-preview | $10.00 | $50.00 | 1M | $1.00 |
| claude-opus-4-1 | $15.00 | $75.00 | 200K | $1.50 |
| claude-opus-4-1-20250805 | $15.00 | $75.00 | 200K | $1.50 |
| claude-opus-4-20250514 | $15.00 | $75.00 | 200K | $1.50 |
| claude-opus-4-5 | $5.00 | $25.00 | 200K | $0.50 |
| claude-opus-4-5-20251101 | $5.00 | $25.00 | 200K | $0.50 |
| claude-opus-4-6 | $5.00 | $25.00 | 1M | $0.50 |
| claude-opus-4-6-20260205 | $5.00 | $25.00 | 1M | $0.50 |
| claude-opus-4-7 | $5.00 | $25.00 | 1M | $0.50 |
| claude-opus-4-7-20260416 | $5.00 | $25.00 | 1M | $0.50 |
| claude-opus-4-8 | $5.00 | $25.00 | 1M | $0.50 |
| claude-opus-5 | $5.00 | $25.00 | 1M | $0.50 |
| claude-sonnet-4-20250514 | $3.00 | $15.00 | 1M | $0.30 |
| claude-sonnet-4-5 | $3.00 | $15.00 | 200K | $0.30 |
| claude-sonnet-4-5-20250929 | $3.00 | $15.00 | 200K | $0.30 |
| claude-sonnet-4-6 | $3.00 | $15.00 | 1M | $0.30 |
| claude-sonnet-5 | $2.00 | $10.00 | 1M | $0.20 |
Prices last synced Aug 20, 2026 from the LiteLLM community price catalog — the same catalog Marginal prices real usage events against, refreshed daily. Spot an error? Tell us.
Other providers
Using Anthropic in production?
Marginal prices every call your app makes against this catalog and slices spend by customer, feature, or any field you define. One track() call to integrate.