OpenAPI
https://api.batchin.tech/openapi.json
Agent entry
This public entry is for autonomous agents, tool runners, and infra teams that need MCP, llms.txt, agents.md, OpenAPI, payment protocol status, and agent-specific access details in one place.
Production posture
Scale
Stable text and multimodal traffic
Protocol set
MCP
llms.txt, agents.md, OpenAPI, JSON-RPC
Traffic mix
Text + MM
One access model across text and multimodal
Funding posture
USDC + Alipay
Separate from ERC-8004 and x402 semantics
OpenAPI
https://api.batchin.tech/openapi.json
API base
https://api.batchin.tech/v1
MCP
https://api.batchin.tech/v1/mcp
llms.txt
https://batchin.tech/llms.txt
llms-full.txt
https://batchin.tech/llms-full.txt
agents.md
https://batchin.tech/agents.md
Agent Skills
https://batchin.tech/.well-known/agent-skills/index.json
MCP Server Card
https://batchin.tech/.well-known/mcp/server-card.json
Agent Card
https://batchin.tech/.well-known/agent-card.json
API Catalog
https://batchin.tech/.well-known/api-catalog
Use for authenticated text and multimodal inference, routing, usage, and billing.
Use for tool discovery and JSON-RPC transport on the stable BatchIn API contract.
Use for run budgets, tool-call traces, retries, request logs, VaaS records, and agent-specific billing visibility.
Reserved inference, dedicated endpoints, and 8+ GPU delivery keep access predictable while isolating delivery and quota.
Payment and proof
Human accounts fund workspaces through the enabled payment rails. Agent-facing proof and settlement protocols run alongside that funding flow.
Supported as a funding rail where settlement is enabled for the workspace.
Chinese self-serve accounts use Alipay where the domestic payment rail is enabled.
Use for receipt and ledger verification. Treat it as proof and audit posture, not a generic payment rail.
Use only where the account is enabled and settlement is live on the English site.
Reliability posture
Next links
Human developers still start in Developer Hub, while agents consume the machine-readable contract. Enterprise and reserved-capacity paths keep the same public truth and diverge only in delivery and entitlement.
Support: hello@luminapath.ai · https://batchin.tech/en/models · https://batchin.tech/en/products