For high-volume, latency-sensitive tasks such as classification, extraction, and routing
Input: 0.10 (≤100k) 0.50 (>100k); Output: 0.50 (≤100k) 2.50 (>100k) per 1M tokens.
| Model | Context | Max out | Pricing |
|---|---|---|---|
| Claude Haiku 5.5 | 1M | 128K | $0.1$0.5 |
| anthropic/claude-haiku-4-5 | 200K | 64K | $1$5 |
/v1/models/perplexity/anthropic/claude-haiku-5-5For high-volume, latency-sensitive tasks such as classification, extraction, and routing