Claude & OpenAI API Pricing
USD per 1M tokens. Pick a supply group per API key: the price is the vendor list price (Anthropic or OpenAI) times the group multiplier. No sign-up required to read prices.
Supply groups
Official · 0.85×
First-party vendor APIs (Anthropic, OpenAI). Most stable.
Bedrock · 0.65×
Routed through AWS Bedrock.
Subscription · 0.3×
Claude Code clients only. Lowest price.
- Official: best for production
- Anthropic's first-party API. The most complete feature set, including server tools, batches and the Files API.
- Bedrock: lower price, fewer features
- Good for chat, coding and tool use. Bedrock does not support server tools (e.g. web search), the Files API, Message Batches or the MCP connector.
- Subscription: Claude Code, best effort
- Lowest price, for Claude Code clients only. Best effort on a shared pool: it may throttle, and we make no uptime promise for this group.
14-day refund: if you buy API credit as a consumer, you may cancel a top-up within 14 days of paying, without giving a reason, and we refund the unused part. Credit already used is not refunded. Refund policy
Claude
| Model | Context | Max output | List price (Anthropic) | Official · 0.85× | Bedrock · 0.65× | Subscription · 0.3× | ||||
|---|---|---|---|---|---|---|---|---|---|---|
| Input | Output | Input | Output | Input | Output | Input | Output | |||
| Claude Fable 5.1 claude-fable-5-1 | 1M | 128K | $10 | $50 | $8.5 | $42.5 | $6.5 | $32.5 | $3 | $15 |
| Claude Fable 5 claude-fable-5 | 1M | 128K | $10 | $50 | $8.5 | $42.5 | $6.5 | $32.5 | $3 | $15 |
| Claude Opus 5.5 claude-opus-5-5 | 1M | 128K | $4 | $20 | $3.4 | $17 | $2.6 | $13 | $1.2 | $6 |
| Claude Opus 5 claude-opus-5 | 1M | 128K | $5 | $25 | $4.25 | $21.25 | $3.25 | $16.25 | $1.5 | $7.5 |
| Claude Opus 4.8 claude-opus-4-8 | 1M | 128K | $5 | $25 | $4.25 | $21.25 | $3.25 | $16.25 | $1.5 | $7.5 |
| Claude Opus 4.7 claude-opus-4-7 | 1M | 128K | $5 | $25 | $4.25 | $21.25 | $3.25 | $16.25 | $1.5 | $7.5 |
| Claude Sonnet 5.5 claude-sonnet-5-5 | 1M | 128K | $2 | $10 | $1.7 | $8.5 | $1.3 | $6.5 | $0.6 | $3 |
| Claude Sonnet 5 claude-sonnet-5 | 1M | 128K | $2 | $10 | $1.7 | $8.5 | $1.3 | $6.5 | $0.6 | $3 |
| Claude Sonnet 4.6 claude-sonnet-4-6 | 1M | 128K | $3 | $15 | $2.55 | $12.75 | $1.95 | $9.75 | $0.9 | $4.5 |
| Claude Haiku 5.5 claude-haiku-5-5 List price applies to prompts up to 100K tokens. | 1M | 128K | $0.1 | $0.5 | $0.085 | $0.425 | $0.065 | $0.325 | $0.03 | $0.15 |
| Claude Haiku 4.5 claude-haiku-4-5 | 200K | 64K | $1 | $5 | $0.85 | $4.25 | $0.65 | $3.25 | $0.3 | $1.5 |
List prices are Anthropic's published API prices (USD per 1M tokens) as of 2026-10-25. The Subscription group works with Claude Code clients only. Cache read/write prices are shown on each model page.
OpenAI
| Model | Context | Max output | List price (OpenAI) | Official · 0.85× | ||
|---|---|---|---|---|---|---|
| Input | Output | Input | Output | |||
| GPT-6 Astra gpt-6-astra | 1.05M | 128K | $10 | $50 | $8.5 | $42.5 |
| GPT-6.1 Sol gpt-6.1-sol | 1.05M | 128K | $2 | $10 | $1.7 | $8.5 |
| GPT-6 Sol gpt-6-sol | — | — | $2 | $10 | $1.7 | $8.5 |
| GPT-6 Luna gpt-6-luna | 1.05M | 128K | $0.1 | $0.5 | $0.085 | $0.425 |
| GPT-5.6 Sol gpt-5.6-sol | — | — | $4 | $20 | $3.4 | $17 |
| GPT-5.6 Terra gpt-5.6-terra | — | — | $2 | $12 | $1.7 | $10.2 |
| GPT-5.6 Luna gpt-5.6-luna | — | — | $0.2 | $1.2 | $0.17 | $1.02 |
| GPT-5.5 gpt-5.5 List price applies to prompts under 272K tokens; longer prompts are billed at a higher long-context rate. | — | — | $5 | $30 | $4.25 | $25.5 |
| GPT-5.4 gpt-5.4 List price applies to prompts under 272K tokens; longer prompts are billed at a higher long-context rate. | — | — | $2.5 | $15 | $2.13 | $12.75 |
| GPT-5.4 mini gpt-5.4-mini | — | — | $0.75 | $4.5 | $0.637 | $3.83 |
| GPT-5.4 nano gpt-5.4-nano | — | — | $0.2 | $1.25 | $0.17 | $1.06 |
| GPT-5.3 Codex gpt-5.3-codex | — | — | $1.75 | $14 | $1.49 | $11.9 |
OpenAI models are served by the Official group only and priced as OpenAI's list price (USD per 1M tokens, checked 2026-10-25) times the group multiplier. Cached input is shown on each model page; OpenAI charges no cache-write surcharge. A dash means OpenAI does not state the figure on its models page. Per-key limits for OpenAI models are the Official group defaults below.
Limits by supply group
Context and max output are per model (synchronous Messages API; checked 2026-10-25). Per-key defaults below.
| Supply group | Requests / min | Tokens / min | Concurrent | Notes |
|---|---|---|---|---|
| Official | 60 | 400K | 8 |
|
| Bedrock | 60 | 400K | 8 |
|
| Subscription Claude Code only · best effort · shared pool | 20 | 150K | 3 |
|
Tokens per minute count uncached input plus output; cache reads are not counted. Limits are per-key defaults, not guarantees. Need higher limits? Contact support@ahrigate.com.
Refunds
14-day refund of the unused part of any top-up (consumers: statutory right of withdrawal; businesses: on request). Refund policy · billing@ahrigate.com