LLM Gateway

A unified interface for LLMs — pay with your launch fees or wallet balance.

Access Claude, Gemini, and GPT models through a single API. Every request is tracked, and you can fund usage directly from your token launch fees or authorize payments in ETH, USDC, or BANKR.

Why Bankr LLM Gateway?

Feature	Benefit
Cost Tracking	Full visibility into token usage and costs per request
Launch Fee Funding	Automatically allocate a portion of your token launch fees to AI
Flexible Payments	Pay with ETH, USDC, or BANKR — your choice
Multi-Provider	Access models from Anthropic, Google, OpenAI, Moonshot AI, and Alibaba
High Availability	Automatic failover ensures your agents stay online
SDK Compatible	Works with OpenAI and Anthropic SDKs — no code changes

Supported Models

Claude (Anthropic)

Model	Context	Input	Best For
claude-opus-4.6	200K	text, image	Most capable, advanced reasoning
claude-opus-4.5	200K	text, image	Complex reasoning, analysis
claude-sonnet-4.5	200K	text, image	Balanced speed and quality
claude-haiku-4.5	200K	text, image	Fast, cost-effective

Gemini (Google)

Model	Context	Input	Best For
gemini-3-pro	2M	text, image	Advanced reasoning
gemini-3-flash	1M	text, image	Fast, high throughput
gemini-2.5-pro	1M	text, image	Long context, multimodal
gemini-2.5-flash	1M	text, image	Speed, high throughput

GPT (OpenAI)

Model	Context	Input	Best For
gpt-5.2	262K	text	Advanced reasoning
gpt-5.2-codex	262K	text	Code generation
gpt-5-mini	128K	text	Fast, economical
gpt-5-nano	128K	text	Ultra-fast, lowest cost

Kimi (Moonshot AI)

Model	Context	Input	Best For
kimi-k2.5	128K	text	Long-context reasoning

Qwen (Alibaba)

Model	Context	Input	Best For
qwen3-coder	128K	text	Code generation, debugging

How It Works

┌─────────────┐  ┌─────────────┐  ┌─────────────┐
│  OpenClaw   │  │ Claude Code │  │ Custom Apps │
└──────┬──────┘  └──────┬──────┘  └──────┬──────┘
       │                │                │
       └────────────────┼────────────────┘
                        │
                        ▼
              ┌─────────────────┐
              │   LLM Gateway   │
              │  llm.bankr.bot  │
              └────────┬────────┘
                       │
        ┌──────────────┼──────────────┐
        ▼              ▼              ▼
   Vertex AI      Vertex AI      OpenRouter
   (Gemini)       (Claude)       (All Models)

Routing Strategy:

• Gemini models → Vertex AI (primary), OpenRouter (fallback)
• Claude models → Vertex AI (primary), OpenRouter (fallback)
• GPT, Kimi, Qwen models → OpenRouter only

Quick Start

Get Your API Key

Get your LLM Gateway API key →

Base URL:

https://llm.bankr.bot

Example Request:

curl https://llm.bankr.bot/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "X-API-Key: bk_YOUR_API_KEY" \
  -d '{
    "model": "claude-opus-4.6",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'

Key Permissions

Your API key must have llmGatewayEnabled turned on at bankr.bot/api.

• LLM Gateway calls are inherently read-only — the readOnly flag does not affect gateway access.
• A separate LLM key only needs llmGatewayEnabled (no agentApiEnabled required).

See Access Control for full details.

Setup with Bankr CLI

The Bankr CLI can configure your tools automatically:

bun install -g @bankr/cli
bankr login
bankr llm setup openclaw --install   # OpenClaw
bankr llm setup opencode --install   # OpenCode
bankr llm setup claude               # Claude Code
bankr llm setup cursor               # Cursor

If your LLM gateway key differs from your Bankr API key, you can set them separately:

bankr login --llm-key YOUR_LLM_KEY
# or
bankr config set llmKey YOUR_LLM_KEY

See AI Coding Tools for full details on each tool.

Next Steps

• API Reference — Full endpoint documentation
• OpenClaw — Configure your OpenClaw agent
• AI Coding Tools — Claude Code, Codex, Cursor, OpenCode, and more
• Install the Bankr Skill — Give your OpenClaw agent crypto and DeFi capabilities