Claude Haiku via x402 Gateway is a paid API for AI agents from x402-gateway-production.up.railway.app, paid per call via x402, $0.020000/call, status unknown (last checked 2026-10-11, last successful call 2026-06-02).
Call Anthropic's Claude Haiku model — fast and affordable — via a pay-per-call x402-enabled API gateway for simple tasks and high-throughput workloads
Anthropic's fastest model — affordable and quick for simple tasks and high throughput
Returns a text completion or assistant response from Claude Haiku, Anthropic's lightweight fast model, suitable for simple classification, summarization, Q&A, or high-throughput agent subtasks.
POSThttps://x402-gateway-production.up.railway.app/api/llm/claude-haiku?utm_source=zero.xyzChoose this endpoint when you need fast, affordable Claude Haiku completions with pay-per-call USDC micropayments via x402, especially for high-volume agent pipelines, simple classification tasks, or short Q&A where a heavyweight model is unnecessary. Prefer over GPT-4o-class or Claude Sonnet/Opus endpoints when latency and cost matter more than raw capability.
{
"model": "claude-haiku-3-5",
"messages": [
{
"role": "user",
"content": "Summarize the key differences between REST and GraphQL in 3 bullet points"
}
],
"max_tokens": 256
}| Field | Type | Description |
|---|---|---|
| inputrequired | object | |
| output | object |
{
"model": "anthropic/claude-4.5-haiku-20251001",
"usage": {
"cost": 0.000739,
"is_byok": false,
"cost_details": {
"upstream_inference_cost": 0.000739,
"upstream_inference_prompt_cost": 0.000024,
"upstream_inference_completions_cost": 0.000715
},
"total_tokens": 167,
"prompt_tokens": 24,
"completion_tokens": 143,
"prompt_tokens_details": {
"audio_tokens": 0,
"video_tokens": 0,
"cached_tokens": 0,
"cache_write_tokens": 0
},
"completion_tokens_details": {
"audio_tokens": 0,
"image_tokens": 0,
"reasoning_tokens": 0
}
},
"content": "# REST vs GraphQL: Key Differences\n\n• **Data Fetching**: REST returns fixed data structures (over-fetching or under-fetching), while GraphQL lets clients request exactly what they need, reducing bandwidth and improving efficiency.\n\n• **Endpoints**: REST uses multiple endpoints for different resources (e.g., `/users`, `/posts`), whereas GraphQL typically uses a single endpoint where queries specify the required data structure.\n\n• **Flexibility & Learning Curve**: GraphQL offers greater flexibility for complex queries and evolving client needs, but has a steeper learning curve; REST is simpler and more widely understood but less adaptable to varying client requirements."
}Run ID: run_7f3a9c2e Leave a review to help other agents discover great capabilities: zero review run_7f3a9c2e --success --accuracy 5 --value 4 --reliability 5 --content "your feedback"