OpenAI-compatible — swap the base URL and it works with your existing SDK.
| Type | Price |
|---|---|
| Input | |
| Output | |
| Cached input |
| Benchmark | Score | What it measures | Source | Retrieved |
|---|---|---|---|---|
| Intelligence | 5.6 | Composite score across standardised reasoning, knowledge and problem-solving evaluations, measured independently by Artificial Analysis | Source | September 12, 2026 |
| Model | Input | Output | Context | Best for |
|---|---|---|---|---|
Claude-Haiku-3 This page | Fast, low-cost tasks | |||
| Reasoning + agents | ||||
| Balanced coding + agents | ||||
| Long-context, multimodal & agentic workflows | ||||
| Reasoning + agents |
Claude-Haiku-3 has a 1,000,000 tokens context window and can return up to 200K.
Use anthropic/claude-3-haiku as the model id. Requests go to v1/chat/completions.
Claude-Haiku-3 is priced at input $0.34385 / 1M tokens, output $1.625 / 1M tokens, cached input $0.041262 / 1M tokens.
Yes, Claude-Haiku-3 can stream responses as they are generated.
Yes, Claude-Haiku-3 supports both function calling and structured outputs.
Claude-Haiku-3 was built by Anthropic.
