{"service":"x402-llm-gateway","description":"Privacy-first chat API: prompts run on a local open-weight model on dedicated hardware - your data is never forwarded to OpenAI/Anthropic or any third-party LLM cloud. OpenAI-compatible chat completions: summarize, draft, extract, translate, code, answer questions. Send a messages array, get chat.completion JSON. Pay per call in USDC on Base (x402), no API key or account. Tiers by output cap: quick 4,096 / long 8,192 / extended 32,768 tokens. Free trial: POST /v1/trial.","what_this_is":"Private, self-hosted LLM inference for AI agents. A local 27B open-weight model (Qwen) with OpenAI-compatible endpoints, plus 768-dim embeddings and image (vision) understanding. Inference runs on this operator's own hardware, so your prompts are NOT sent to a third-party cloud API - the right choice where data can't go to OpenAI/Anthropic, or where you want predictable per-call pricing.","why_use_us":["Private: prompts are processed locally and not forwarded to a big-cloud LLM API.","No API key, no account, no signup - pay per call in USDC (x402), gasless for the payer.","Predictable per-call price with no provider-side rate limits and no surprise per-token bills.","Includes cheap 768-dim embeddings (the high-frequency RAG/retrieval call) for far less than a chat call.","Image understanding: point the local vision-language model at a public image URL and get a private, paid answer (vision tier)."],"getting_started":{"no_wallet_or_free":"No USDC wallet yet? Use the free trial (open during promo windows): POST /v1/trial with a messages array. About 10 calls/hour per IP.","to_pay":["1) GET this card (you are here) to learn the tiers, prices and endpoints.","2) POST to a paid endpoint WITHOUT payment -> you get HTTP 402 with a PAYMENT-REQUIRED header.","3) Sign the USDC authorization (EIP-3009) with your Base wallet; retry the same request with the PAYMENT-SIGNATURE header.","4) You receive the OpenAI chat.completion JSON. The x402 facilitator settles the USDC; you pay no gas."],"example_free_trial":{"method":"POST","path":"/v1/trial","body":{"model":"qwen/qwen3.8-27b","messages":[{"role":"user","content":"Classify into billing|bug|feature|other: 'My invoice never arrived and I need it for accounting.'"}]}},"example_paid_quick":{"method":"POST","path":"/v1/chat/completions","body":{"model":"qwen/qwen3.8-27b","messages":[{"role":"user","content":"Summarize in one sentence: <paste text>"}],"max_tokens":1024}},"example_embeddings":{"method":"POST","path":"/v1/embeddings","body":{"model":"text-embedding-nomic-embed-text-v1.5","input":["text one","text two"]}},"example_vision":{"method":"POST","path":"/v1/vision","body":{"model":"qwen/qwen3.8-27b","image":"https://example.com/photo.jpg","prompt":"What is in this image?"}}},"tiers":[{"tier":"quick","path":"/v1/chat/completions","price":"$0.001","best_for":"Classification, extraction, short Q&A, tool calls, one-shot agent steps.","output_cap_tokens":4096},{"tier":"long","path":"/v1/long","price":"$0.05","best_for":"Summarization, drafting, multi-paragraph answers.","output_cap_tokens":8192},{"tier":"extended","path":"/v1/extended","price":"$0.20","best_for":"Long-form writing, reports, full document drafts.","output_cap_tokens":32768},{"tier":"embed","path":"/v1/embeddings","price":"$0.001","best_for":"RAG, semantic search, dedupe, clustering - the call you make hundreds of times.","output":"768-dim vectors, up to 64 texts/call"},{"tier":"vision","path":"/v1/vision","price":"$0.02","best_for":"Image understanding - describe/OCR/classify a public image URL. Up to 4 images per call.","output":"chat.completion JSON, 2,048-token cap"}],"embeddings":"High-frequency workhorse: 768-dim vectors (nomic-embed-text-v1.5) for $0.001/call, up to 64 texts. This is the cheap, predictable call for retrieval/RAG/dedupe in agent pipelines - use it far more than the chat tiers.","vision":"Image understanding from the same local vision-language model: send a public image URL + a prompt for $0.02/call, up to 4 images, 2,048-token answers. Private - the image and your prompt never touch a third-party vision API.","paid_endpoints":{"POST /v1/chat/completions":{"tier":"quick","price":"$0.001","output_cap_tokens":4096,"network":"eip155:8453","pay_to":"0x23976FB6Af8f7fE8756730f0473fB7FfaC531ee8","scheme":"exact"},"POST /v1/long":{"tier":"long","price":"$0.05","output_cap_tokens":8192,"network":"eip155:8453","pay_to":"0x23976FB6Af8f7fE8756730f0473fB7FfaC531ee8","scheme":"exact"},"POST /v1/extended":{"tier":"extended","price":"$0.20","output_cap_tokens":32768,"network":"eip155:8453","pay_to":"0x23976FB6Af8f7fE8756730f0473fB7FfaC531ee8","scheme":"exact"},"POST /v1/embeddings":{"tier":"embed","price":"$0.001","model":"text-embedding-nomic-embed-text-v1.5","max_batch_texts":64,"network":"eip155:8453","pay_to":"0x23976FB6Af8f7fE8756730f0473fB7FfaC531ee8","scheme":"exact"},"POST /v1/vision":{"tier":"vision","price":"$0.02","model":"qwen/qwen3.8-27b","max_images":4,"output_cap_tokens":2048,"network":"eip155:8453","pay_to":"0x23976FB6Af8f7fE8756730f0473fB7FfaC531ee8","scheme":"exact"}},"free_endpoints":["GET /v1/models","POST /v1/trial (free trial - promo window OPEN)"],"trial":{"enabled":true,"since":1789709080955,"path":"/v1/trial","output_cap_tokens":2048,"rate_per_hour":10},"protocol":"x402 (HTTP 402 Payment Required)","facilitator":"https://api.cdp.coinbase.com/platform/v2/x402"}