Input
- Official Anthropic
- $1
- Here, from (−60%)
- $0.4
- Here, best (−80%)
- $0.2
Claude Haiku 4.5 is the fastest and cheapest Claude model — built for high-volume, latency-sensitive work like classification, extraction and routing.
| Rate | Official Anthropic | Here, from (−60%) | Here, best (−80%) |
|---|---|---|---|
| Input | $1 | $0.4 | $0.2 |
| Output | $5 | $2 | $1 |
| Cache read | $0.1 | $0.04 | $0.02 |
| Cache write (5m) | $1.25 | $0.5 | $0.25 |
Every request is metered at the official rate first, then your progressive B2C discount (60% at the start, up to 80% as cumulative top-ups grow) is subtracted before it touches your prepaid balance. Context window: 200K tokens. Max output: 64K tokens.
Create a free account, generate one key, and point any Anthropic-compatible tool at https://api.apitoken.sale with model ID claude-haiku-4-5. New accounts include $10 of Claude usage at official API prices — enough to test the model before topping up.
Officially $1 per 1M input tokens and $5 per 1M output tokens. With the apiToken.sale discount that starts at $0.40/$2 and reaches $0.20/$1 — the cheapest way to run Claude.
High-volume, low-latency work: classification, extraction, summarization, routing and simple chat. For complex reasoning, step up to Sonnet 5 or Opus 4.8.
claude-haiku-4-5. It works on the same apiToken.sale key and balance as every other supported Claude model.
Run Claude Haiku 4.5 on the same Anthropic API at up to 80% off — instant key, prepaid balance, card or crypto.