# BridgeNode

> AI inference bridge for AI agents. No API keys. Free models included, plus
> free trials on paid models — pay with Solana USDC via x402 after that.

Full agent guide: https://bridgenode.cc/llms-full.txt

## Free access (no wallet, no API key, no registration)

- Free models: gpt-oss-20b, gpt-oss-120b, glm-4.7-flash, glm-4.5-flash, glm-4.6v-flash
**Free model notes (read before choosing one):**
- `glm-4.7-flash` — ⚠️ **temporarily unreliable**: z.ai free model: slower than the Groq free models — a reply can take up to a minute, and the provider is sometimes overloaded. If it returns an error (rate limit / temporarily overloaded), retry once or switch to gpt-oss-20b, the most reliable free model.
- `glm-4.5-flash` — ⚠️ **temporarily unreliable**: z.ai free model: slower than the Groq free models — a reply can take up to a minute, and the provider is sometimes overloaded. If it returns an error (rate limit / temporarily overloaded), retry once or switch to gpt-oss-20b, the most reliable free model.
- `glm-4.6v-flash`: z.ai free model: slower than the Groq free models — a reply can take up to a minute, and the provider is sometimes overloaded. If it returns an error (rate limit / temporarily overloaded), retry once or switch to gpt-oss-20b, the most reliable free model.
- Free trials: the first 2 call(s) to ANY PAID model are free per client
  (no wallet needed) — so you can try a real model before paying. Trial responses
  carry the headers `X-Bridgenode-Free-Trial: 1` and
  `X-Bridgenode-Free-Trials-Remaining: <n>`.
- Every 402 also carries `extensions.bridgenode`: the free model list, the trials
  left, and links to the payment docs.
- List them live: `GET https://bridgenode.cc/v1/models` (free models have pricing 0 and `"free": true`).

## OpenAI-compatible endpoint
- URL: https://bridgenode.cc/v1
- Models: gpt-oss-20b, gpt-oss-120b, glm-4.7-flash, glm-4.5-flash, glm-4.6v-flash, deepseek-flash, glm-4.7-flashx, glm-5.2, glm-5.1, glm-5, glm-5-turbo, glm-4.7, glm-4.6, glm-4.5, glm-4.5-x, glm-4.5-air, glm-4.5-airx, glm-4-32b-0414-128k, glm-5v-turbo, glm-4.6v, glm-4.6v-flashx, glm-4.5v, kimi-k2.7-code, kimi-k2.7-code-highspeed, kimi-k2.6, MiniMax-M2.7, MiniMax-M2.7-highspeed, MiniMax-M2.5, MiniMax-M2.5-highspeed, MiniMax-M2.1, MiniMax-M2.1-highspeed, MiniMax-M2, deepseek-v4-pro, kimi-k3, glm-5.3, minimax-m3
- Authentication: x402 (PAYMENT-SIGNATURE header)
- Network: Solana mainnet (USDC)
- Gasless: BridgeNode covers transaction fees (agent needs 0 SOL)

## MCP endpoint
- URL: https://bridgenode.cc/mcp
- Transport: Streamable HTTP (only; HTTP+SSE is deprecated, SEP-2596)
- Authentication: x402 handshake
- Install: `npm i @bridgenode/mcp` (https://www.npmjs.com/package/@bridgenode/mcp)

## Tool calling (function calling)
- Send OpenAI-style `tools` (+ `tool_choice`) with the request; the answer is the
  provider's own (`choices[0].message.tool_calls`, `finish_reason: "tool_calls"`).
- The tool schema counts as input tokens (priced and context-checked); models
  marked `"tools": true` in `GET https://bridgenode.cc/v1/models` are verified to accept tools.
- Full contract + the agent loop: https://bridgenode.cc/skill.md

## Supported clients
- Any OpenAI-compatible client with x402 support
- Claude Code, Codex, Cursor, Windsurf
- Solana `pay` CLI (native Solana x402 client)
- MCP-compatible agents

## Agent skill
- Install: `npx skills add bridgenode-ai/bridgenode-skill`
- Source: https://github.com/bridgenode-ai/bridgenode-skill
- Skill file: https://bridgenode.cc/skill.md

## SDKs
- Python: `pip install bridgenode-llm` (https://pypi.org/project/bridgenode-llm)
- TypeScript: `npm i @bridgenode/llm` (https://www.npmjs.com/package/@bridgenode/llm)

## First call (free — copy this)
# No wallet, no 402: the free model answers directly. max_tokens >= 200 —
# a smaller limit can be eaten by reasoning and return an empty answer.
curl https://bridgenode.cc/v1/chat/completions \
  -H "Content-Type: application/json" \
  -d '{"model":"gpt-oss-20b","messages":[{"role":"user","content":"hello"}],"max_tokens":200}'

## Quick start (x402 V2 flow: request → 402 → sign → retry)
# 1) Request — response: 402 PAYMENT-REQUIRED (price, payTo, memo)
curl https://bridgenode.cc/v1/chat/completions \
  -H "Content-Type: application/json" \
  -d '{"model":"deepseek-flash","messages":[{"role":"user","content":"hello"}],"max_tokens":200}'

# 2) Sign the partial TX (USDC TransferChecked + Memo) and retry with header:
#    PAYMENT-SIGNATURE: <base64 PaymentPayload — full flow: https://bridgenode.cc/skill.md>
curl https://bridgenode.cc/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "PAYMENT-SIGNATURE: <base64 PaymentPayload>" \
  -d '{"model":"deepseek-flash","messages":[{"role":"user","content":"hello"}],"max_tokens":200}'

## Pricing
- Prices: https://bridgenode.cc/models — live prices: https://bridgenode.cc/v1/models
- Reasoning/thinking models: use `max_tokens >= 200` (reasoning tokens share the `max_tokens` budget; a too-small limit can produce an EMPTY answer — we retry once with a bigger budget and refund in full if it stays empty). Thinking is disabled on: `glm-4.7-flash`, `deepseek-flash`, `deepseek-v4-pro`.

## Community
- Listed on x402-list: https://x402-list.com/services/bridgenode
- Listed on x402-dev: https://www.x402dev.com/awesome-projects/
- Listed on nohumans.directory: https://nohumans.directory/l/f1f74751-9d5
- Listed on gold-402: https://github.com/Haustorium12/gold-402/blob/main/directory/learning.md
- ClawHub skill: https://clawhub.ai/bridgenode/skills/bridgenode
- Moltbook agent profile: https://moltbook.com/u/bridgenode_eli
- Discord: https://discord.gg/HUpVKxJxhG
- X (Twitter): https://x.com/eliBNx