Input
- Official Anthropic
- $0.1
- Here (−50%)
- $0.05
claude-haiku-5-5
Claude Haiku 5.5 API — price per token
Claude Haiku 5.5 is Anthropic's October 2026 Haiku and the cheapest, fastest Claude model, with a 1M context window, adaptive thinking and effort control.
Pricing per 1M tokens
| Rate | Official Anthropic | Here (−50%) |
|---|---|---|
| Input | $0.1 | $0.05 |
| Output | $0.5 | $0.25 |
| Cache read | $0.01 | $0.005 |
| Cache write (5m) | $0.125 | $0.0625 |
Output
- Official Anthropic
- $0.5
- Here (−50%)
- $0.25
Cache read
- Official Anthropic
- $0.01
- Here (−50%)
- $0.005
Cache write (5m)
- Official Anthropic
- $0.125
- Here (−50%)
- $0.0625
Every request is metered at the official rate first, then the flat 50% B2C discount is subtracted before it touches your prepaid balance. Context window: 1M tokens. Max output: 128K tokens.
Best for
- Classification, extraction, routing and subagent calls at very high volume.
- Latency-sensitive chat and tool loops that need adaptive thinking.
- Cheap pre-processing before an Opus or Sonnet call.
Good to know
- Official rates per 1M tokens for prompts up to 100K tokens: $0.10 input, $0.01 cache read, $0.125 5-minute cache write, $0.20 1-hour cache write, $0.50 output.
- Above 100K prompt tokens (input plus cache reads and writes), the whole request uses $0.50 input, $0.05 cache read, $0.625 / $1 cache writes and $2.50 output per 1M.
- Adaptive thinking is on by default; effort low through max is supported, and thinking type "disabled" works at high effort or below. Manual budget_tokens and a non-default temperature or top_p return 400.
How to use Claude Haiku 5.5
Create a free account, generate one key, and point any Anthropic-compatible tool at https://router.apitoken.sale with model ID claude-haiku-5-5. Eligible new accounts include $5 of platform bonus credit — enough to test the model before topping up. Existing integrations on the legacy host https://api.apitoken.sale keep working.
Frequently asked questions
How much does the Claude Haiku 5.5 API cost?
Officially $0.10 per 1M input and $0.50 per 1M output tokens for prompts up to 100K tokens, and $0.50/$2.50 for the whole request above that. apiToken.sale applies the flat 50% discount: $0.05/$0.25 (or $0.25/$1.25 for long prompts).
What is the model ID for Claude Haiku 5.5?
claude-haiku-5-5. Use it unchanged with the Anthropic SDK, Claude Code, Cursor, Cline or any compatible tool pointed at https://router.apitoken.sale.
Haiku 5.5 or Haiku 4.5?
Haiku 5.5 is newer, ten times cheaper per token for prompts up to 100K tokens, and has a 1M context window. Code written for Haiku 4.5 must drop budget_tokens thinking and non-default sampling parameters, and read content blocks by type, because responses can begin with thinking.
Run Claude Haiku 5.5 on the same Anthropic API at a flat 50% off — instant key, prepaid balance, card or crypto.