AgentDrop Inference

fyi.agentdropPublished by agentdrop.fyiOfficial publisherv0.3.0Updated Nov 1, 2026

Pay-per-call chat on GPT-6.1, Claude, Gemini and Grok over x402. No API key; a failed call is free.

The publisher's own server: AgentDrop

VerifiedStreamable HTTPWeb executableFinanceAI & ML

Overview

AI-generated overview

Remote MCP endpoint for pay-per-call chat completions on frontier models (Claude, GPT, Gemini, Grok), billed in USDC over x402 with no API key.

What you can do
  • Run chat completions on frontier models such as Claude Sonnet/Haiku, GPT-6.1, Gemini and Grok, using an OpenAI-style messages request with optional max_tokens and temperature.
  • List available models together with provider rates, the pricing formula and limits, free of charge.
  • Get a per-request price quote in the 402 response before a call runs, and pay only for calls that succeed.
When to use it
Useful when an assistant needs occasional access to several frontier models without signing up for separate provider accounts or managing API keys, and when paying per call from a wallet is acceptable.
Requirements
A remote streamable HTTP endpoint; no package, account or API key is declared. Calls are paid in USDC over x402, so a funded wallet capable of x402 payments is needed; the paying wallet serves as the identity. Network access to the endpoint is required.
Before you install
Every chat call spends real money in USDC from the paying wallet; the price is quoted in the 402 before execution, so review it. There is no account or key to revoke, so wallet control is the only safeguard. Prompts are sent to the model's provider and are not stored by the service. Thinking models spend part of max_tokens on reasoning.

Installation

In SourceWeft

  1. Open AgentDrop Inference in the dashboard and add it to a workspace.
  2. Enable the server for the chats that should use its tools.

Web executable via Streamable HTTP. Remote servers run from the web runtime once configured in a workspace.

Use with an agent

Paste this prompt into Claude Code, Codex, Cursor or another agent — or into SourceWeft chat. The agent reads this server's install guide, shows you its publisher, endpoint and sign-in, and adds it to your client once you agree. Keys never go into a command.

Read https://sourceweft.com/mcp/fyi.agentdrop%2Fai/install.md and connect the MCP server it describes to my client. Before changing any config, show me its publisher, endpoint and sign-in method and wait for my OK. Never put an API key in a command or in a file I would commit; ask me where it should come from.

Read the install guide the agent follows

Other MCP clients

Add this to your client's mcpServers config.

{
  "mcpServers": {
    "agentdrop-inference": {
      "type": "http",
      "url": "https://agentdrop.fyi/ai/mcp"
    }
  }
}

README

This repository has no README.

Tools

2

chat

Chat

Chat completion on a frontier model: GPT-6.1 Sol and Terra, GPT-5.5, Claude Sonnet 5.5 and Haiku 5.5, Gemini 3.8 Flash, Grok 4.7, GPT-6 Luna (default anthropic/claude-haiku-5.5; see the models tool). Same shape as OpenAI chat: messages, optional max_tokens (default 1024, max 8192), temperature, tools. Price is quoted per call from your input size and max_tokens: (estimated input tokens x input rate + max_tokens x output rate) x 1.4 + $0.0005, minimum $0.002; the 402 states the exact amount. Charged only if the model call succeeds. Thinking models (GPT-5.5, Gemini, Grok, gpt-oss) spend part of max_tokens on reasoning: give them room, or set reasoning_effort (default low). Prompts go to the model's provider and are not stored by Payperwork. No streaming.

Parametersstopmodeltoolstop_pmessagesrequiredmax_tokenstemperaturetool_choiceresponse_formatreasoning_effort

models

Models

List the available models with provider rates, the pricing formula and limits. Free.

Takes no parameters.

Version history

1
  1. v0.3.0LatestNov 1, 2026

Score

75% of the applicable checks met

How complete this listing is, computed from the registry entry, the endpoint and public sources. Not a rating of the server and not a statement that it is safe or working today: reachability is a dated check, and no installs or reviews go into the score.

  • Reached in the last 30 daysRequired25 pts
  • Auth declaredRequired10 pts
  • Tools listedRequired15 pts
  • DescriptionRequired15 pts
  • Logo10 pts
  • Documentation15 pts

    Declare a GitHub repository with a README, or serve resource_documentation in the OAuth protected-resource metadata.

  • License10 pts

    Declare a license in the repository or the registry entry.

Other Agentdrop integrations