Model catalogue / Anthropic
Claude Haiku 4.5
Claude Haiku 4.5 on VipAI costs $0.34 per 1M input tokens and $1.70 per 1M output — 66% off the provider's published rate. It has a 200K context window.
$0.34Input / 1M tokens
$1.70Output / 1M tokens
—Cache read / 1M tokens
200KContext window
Price against list
The list column is the provider's own published rate, so the discount can be checked rather than taken on trust.
| Rate | Provider list | VipAI | Discount |
|---|---|---|---|
| Input / 1M tokens | $1.00 | $0.34 | 66% off |
| Output / 1M tokens | $5.00 | $1.70 | — |
| Cache read / 1M tokens | — | — | — |
Endpoints it answers on
The router does not translate protocols silently, so use the shape the family expects. The full matrix is in models & routing.
| Protocol | Method | Path |
|---|---|---|
| OpenAI Chat Completions | POST | /v1/chat/completions |
Calling it
curl https://api.vipai.site/v1/chat/completions \
-H "Authorization: Bearer $VIPAI_API_KEY" \
-H "Content-Type: application/json" \
-d '{ "model": "claude-haiku-4-5", "messages": [{ "role": "user", "content": "Ship it." }] }'Use it
- Create an API key — it works with every model on the key, not just this one.
- API integration — headers, bodies and streaming for each protocol.
- Billing questions — why one message can produce several billing rows.
- All model prices — the full catalogue against list.
Prices are live and move with upstream costs. This page reflects the rate in effect now.