Claude 3.5 Haiku API pricing
RetiredAs of October 22, 2026, Anthropic Claude 3.5 Haiku costs $0.80 per million input tokens and $4.00 per million output tokens as listed on OpenRouter.
Claude 3.5 Haiku benchmarks and specs · Every Anthropic model's price
No longer sold through the API. Retired on February 19, 2026; last list price shown.
- Input, per million tokens$0.80
- Output, per million tokens$4.00
- Cached input, per millionNo cache price listed
- Context window200K tokens
- Max outputNot published
- Output speedNot measured
What a request costs
Three fixed token mixes at the list price, without cache or batch discounts. For a whole month, use the calculator.
| Workload | Tokens | One request | 1,000 requests | Fits the context window |
|---|---|---|---|---|
| Chat turn | 1K input + 500 output tokens | $0.0028 | $2.80 | Yes |
| Repository review | 50K input + 3K output tokens | $0.052 | $52.00 | Yes |
| Agent loop | 200K input + 10K output tokens | $0.200 | $200.00 | No, it would need splitting |
| Solved coding task | One SWE-bench Verified attempt, divided by the share that succeed | Not computable | Not computable |
Price history
The price of Claude 3.5 Haiku here is the price OpenRouter lists, so we do not track price changes for it.
Cheaper models as good as Claude 3.5 Haiku
Models within two points of Claude 3.5 Haiku or above it on the LLMBenchmark Index, with a lower output price. Cheapest first.
- gpt-oss-20b$0.15 Overall 38.5 against 28.2 · 96% cheaper output
- Gemma 3 12B$0.1625 Overall 28.0 against 28.2 · 96% cheaper output
- Qwen3.5-Flash$0.26 Overall 50.1 against 28.2 · 94% cheaper output
- MiMo-V2.6-Flash$0.28 Overall 60.8 against 28.2 · 93% cheaper output
- Qwen3 30B A3B Instruct 2507$0.30 Overall 41.5 against 28.2 · 93% cheaper output
Claude 3.5 Haiku pricing questions
- How much does Claude 3.5 Haiku cost per million tokens?
- $0.80 per million input tokens and $4.00 per million output tokens, as listed on OpenRouter on October 22, 2026.
- Does Claude 3.5 Haiku support prompt caching?
- OpenRouter lists no cached input price for this model, so assume every input token is billed at the full rate.
- Is there a cheaper model as good as Claude 3.5 Haiku?
- gpt-oss-20b scores 38.5 on the LLMBenchmark Index, within two points of Claude 3.5 Haiku or above it, and costs $0.15 per million output tokens.
- Has the price of Claude 3.5 Haiku changed?
- The price of Claude 3.5 Haiku here is the price OpenRouter lists, so we do not track price changes for it.