Get account tokens
VisitGet tokensto see your balance and reward rules. Models share one balance; both input and output count.
Start with your first line of code
Familiar requests. A simpler way to connect.
The website is available for preview; model access is in private testing. New accounts have zero balance and rewarded ads are unavailable.
Create your API key
Go to API Keys to create a key for your app. Keep keys on a server or trusted client, never in a public repository.
Make your first request
Choose a protocol and copy the example. OpenAI base URLs include /v1; Anthropic uses the site root. Both use the same key.
curl "https://token-faucet.top/v1/chat/completions" \
-H "Authorization: Bearer $TOKEN_FAUCET_API_KEY" \
-H "Content-Type: application/json" \
-d '{"model":"kimi-k2.6","messages":[{"role":"user","content":"Hello, world!"}]}'Use your coding assistant
Configuration reference · Full client workflows not yet verified
Requests require at least 116,000 available tokens for a reservation; only actual usage is charged. New accounts have no starter credits and rewarded ads are not available.
OpenCode: one provider, all 16 models
Check opencode --version. Merge the matching configuration into your project’s opencode.json and keep existing providers. Set TOKEN_FAUCET_API_KEY in the terminal that starts OpenCode. In v1, /connect → Other can also store a key under token-faucet. Use /models to select a model in the Token Faucet provider group.
v1 uses provider/npm/options; v2 uses providers/package/settings. Do not combine these formats. token-faucet is the provider ID; the part after the slash is the model ID.
export TOKEN_FAUCET_API_KEY="YOUR_TOKEN_FAUCET_KEY" opencode run --model token-faucet/kimi-k2.6 "Reply only OK"
OpenCode v1 · View the complete model configuration
{
"$schema": "https://opencode.ai/config.json",
"model": "token-faucet/kimi-k2.6",
"provider": {
"token-faucet": {
"name": "Token Faucet",
"npm": "@ai-sdk/openai-compatible",
"options": {
"baseURL": "https://token-faucet.top/v1",
"apiKey": "{env:TOKEN_FAUCET_API_KEY}"
},
"models": {
"kimi-k2.6": {
"name": "Kimi K2.6",
"attachment": false,
"reasoning": false,
"temperature": false,
"tool_call": true,
"modalities": {
"input": [
"text"
],
"output": [
"text"
]
},
"limit": {
"context": 100000,
"output": 16000
}
},
"kimi-k2": {
"name": "Kimi K2",
"attachment": false,
"reasoning": false,
"temperature": false,
"tool_call": true,
"modalities": {
"input": [
"text"
],
"output": [
"text"
]
},
"limit": {
"context": 100000,
"output": 16000
}
},
"DeepSeek-V4-Pro": {
"name": "DeepSeek V4 Pro",
"attachment": false,
"reasoning": false,
"temperature": false,
"tool_call": true,
"modalities": {
"input": [
"text"
],
"output": [
"text"
]
},
"limit": {
"context": 100000,
"output": 16000
}
},
"DeepSeek-V4-Flash": {
"name": "DeepSeek V4 Flash",
"attachment": false,
"reasoning": false,
"temperature": false,
"tool_call": true,
"modalities": {
"input": [
"text"
],
"output": [
"text"
]
},
"limit": {
"context": 100000,
"output": 16000
}
},
"glm-5.1": {
"name": "GLM 5.1",
"attachment": false,
"reasoning": false,
"temperature": false,
"tool_call": true,
"modalities": {
"input": [
"text"
],
"output": [
"text"
]
},
"limit": {
"context": 100000,
"output": 16000
}
},
"glm-5": {
"name": "GLM 5",
"attachment": false,
"reasoning": false,
"temperature": false,
"tool_call": true,
"modalities": {
"input": [
"text"
],
"output": [
"text"
]
},
"limit": {
"context": 100000,
"output": 16000
}
},
"glm-5v-turbo": {
"name": "GLM 5V Turbo",
"attachment": false,
"reasoning": false,
"temperature": false,
"tool_call": true,
"modalities": {
"input": [
"text"
],
"output": [
"text"
]
},
"limit": {
"context": 100000,
"output": 16000
}
},
"qwen-3.6-plus": {
"name": "Qwen 3.6 Plus",
"attachment": false,
"reasoning": false,
"temperature": false,
"tool_call": true,
"modalities": {
"input": [
"text"
],
"output": [
"text"
]
},
"limit": {
"context": 100000,
"output": 16000
}
},
"qwen-3.5": {
"name": "Qwen 3.5",
"attachment": false,
"reasoning": false,
"temperature": false,
"tool_call": true,
"modalities": {
"input": [
"text"
],
"output": [
"text"
]
},
"limit": {
"context": 100000,
"output": 16000
}
},
"minimax-m2.7": {
"name": "MiniMax M2.7",
"attachment": false,
"reasoning": false,
"temperature": false,
"tool_call": true,
"modalities": {
"input": [
"text"
],
"output": [
"text"
]
},
"limit": {
"context": 100000,
"output": 16000
}
},
"minimax-m2.5": {
"name": "MiniMax M2.5",
"attachment": false,
"reasoning": false,
"temperature": false,
"tool_call": true,
"modalities": {
"input": [
"text"
],
"output": [
"text"
]
},
"limit": {
"context": 100000,
"output": 16000
}
},
"Doubao-Seed-2.0-Code": {
"name": "Doubao Seed 2.0 Code",
"attachment": false,
"reasoning": false,
"temperature": false,
"tool_call": true,
"modalities": {
"input": [
"text"
],
"output": [
"text"
]
},
"limit": {
"context": 100000,
"output": 16000
}
},
"Doubao_1_6": {
"name": "Doubao 1.6",
"attachment": false,
"reasoning": false,
"temperature": false,
"tool_call": true,
"modalities": {
"input": [
"text"
],
"output": [
"text"
]
},
"limit": {
"context": 100000,
"output": 16000
}
},
"Doubao-Seed-Code": {
"name": "Doubao Seed Code",
"attachment": false,
"reasoning": false,
"temperature": false,
"tool_call": true,
"modalities": {
"input": [
"text"
],
"output": [
"text"
]
},
"limit": {
"context": 100000,
"output": 16000
}
},
"gemini-2.5-pro-latest": {
"name": "Gemini 2.5 Pro",
"attachment": false,
"reasoning": false,
"temperature": false,
"tool_call": true,
"modalities": {
"input": [
"text"
],
"output": [
"text"
]
},
"limit": {
"context": 100000,
"output": 16000
}
},
"gemini_2.5_flash": {
"name": "Gemini 2.5 Flash",
"attachment": false,
"reasoning": false,
"temperature": false,
"tool_call": true,
"modalities": {
"input": [
"text"
],
"output": [
"text"
]
},
"limit": {
"context": 100000,
"output": 16000
}
}
}
}
}
}OpenCode v2 · View the complete model configuration
{
"$schema": "https://opencode.ai/config.json",
"model": "token-faucet/kimi-k2.6",
"providers": {
"token-faucet": {
"name": "Token Faucet",
"package": "@opencode/ai/providers/openai-compatible",
"env": [
"TOKEN_FAUCET_API_KEY"
],
"settings": {
"baseURL": "https://token-faucet.top/v1"
},
"models": {
"kimi-k2.6": {
"name": "Kimi K2.6",
"family": "Kimi",
"capabilities": {
"tools": true,
"input": [
"text"
],
"output": [
"text"
]
},
"limit": {
"context": 100000,
"output": 16000
}
},
"kimi-k2": {
"name": "Kimi K2",
"family": "Kimi",
"capabilities": {
"tools": true,
"input": [
"text"
],
"output": [
"text"
]
},
"limit": {
"context": 100000,
"output": 16000
}
},
"DeepSeek-V4-Pro": {
"name": "DeepSeek V4 Pro",
"family": "DeepSeek",
"capabilities": {
"tools": true,
"input": [
"text"
],
"output": [
"text"
]
},
"limit": {
"context": 100000,
"output": 16000
}
},
"DeepSeek-V4-Flash": {
"name": "DeepSeek V4 Flash",
"family": "DeepSeek",
"capabilities": {
"tools": true,
"input": [
"text"
],
"output": [
"text"
]
},
"limit": {
"context": 100000,
"output": 16000
}
},
"glm-5.1": {
"name": "GLM 5.1",
"family": "GLM",
"capabilities": {
"tools": true,
"input": [
"text"
],
"output": [
"text"
]
},
"limit": {
"context": 100000,
"output": 16000
}
},
"glm-5": {
"name": "GLM 5",
"family": "GLM",
"capabilities": {
"tools": true,
"input": [
"text"
],
"output": [
"text"
]
},
"limit": {
"context": 100000,
"output": 16000
}
},
"glm-5v-turbo": {
"name": "GLM 5V Turbo",
"family": "GLM",
"capabilities": {
"tools": true,
"input": [
"text"
],
"output": [
"text"
]
},
"limit": {
"context": 100000,
"output": 16000
}
},
"qwen-3.6-plus": {
"name": "Qwen 3.6 Plus",
"family": "Qwen",
"capabilities": {
"tools": true,
"input": [
"text"
],
"output": [
"text"
]
},
"limit": {
"context": 100000,
"output": 16000
}
},
"qwen-3.5": {
"name": "Qwen 3.5",
"family": "Qwen",
"capabilities": {
"tools": true,
"input": [
"text"
],
"output": [
"text"
]
},
"limit": {
"context": 100000,
"output": 16000
}
},
"minimax-m2.7": {
"name": "MiniMax M2.7",
"family": "MiniMax",
"capabilities": {
"tools": true,
"input": [
"text"
],
"output": [
"text"
]
},
"limit": {
"context": 100000,
"output": 16000
}
},
"minimax-m2.5": {
"name": "MiniMax M2.5",
"family": "MiniMax",
"capabilities": {
"tools": true,
"input": [
"text"
],
"output": [
"text"
]
},
"limit": {
"context": 100000,
"output": 16000
}
},
"Doubao-Seed-2.0-Code": {
"name": "Doubao Seed 2.0 Code",
"family": "Doubao",
"capabilities": {
"tools": true,
"input": [
"text"
],
"output": [
"text"
]
},
"limit": {
"context": 100000,
"output": 16000
}
},
"Doubao_1_6": {
"name": "Doubao 1.6",
"family": "Doubao",
"capabilities": {
"tools": true,
"input": [
"text"
],
"output": [
"text"
]
},
"limit": {
"context": 100000,
"output": 16000
}
},
"Doubao-Seed-Code": {
"name": "Doubao Seed Code",
"family": "Doubao",
"capabilities": {
"tools": true,
"input": [
"text"
],
"output": [
"text"
]
},
"limit": {
"context": 100000,
"output": 16000
}
},
"gemini-2.5-pro-latest": {
"name": "Gemini 2.5 Pro",
"family": "Gemini",
"capabilities": {
"tools": true,
"input": [
"text"
],
"output": [
"text"
]
},
"limit": {
"context": 100000,
"output": 16000
}
},
"gemini_2.5_flash": {
"name": "Gemini 2.5 Flash",
"family": "Gemini",
"capabilities": {
"tools": true,
"input": [
"text"
],
"output": [
"text"
]
},
"limit": {
"context": 100000,
"output": 16000
}
}
}
}
}
}Cline: OpenAI Compatible
Choose OpenAI Compatible in settings, then enter the base URL, your Token Faucet key and a model ID below. Set context to 100000 and output to 16000. Keep images, Computer Use, extended thinking and prompt caching off. If your version cannot omit unsupported parameters, use the basic API examples. Full client compatibility is not yet certified.
Base URL: https://token-faucet.top/v1 API Key: YOUR_TOKEN_FAUCET_KEY Model ID: kimi-k2.6 Context: 100000 Max output: 16000 Images: OFFOfficial setup documentation
Aider: custom OpenAI endpoint
Set the environment variables below only in the terminal running Aider. Prefix model names with openai/. Configure unknown-model context and output limits to match this service. Aider may send sampling parameters; inspect 400 errors instead of repeatedly retrying.
export OPENAI_API_BASE="https://token-faucet.top/v1" export OPENAI_API_KEY="YOUR_TOKEN_FAUCET_KEY" aider --model openai/kimi-k2.6Official setup documentation
Current compatibility limits
Use Chat Completions, not Responses. Configure every model for text input; native image capability does not mean images work through this service. Omit max_tokens or use 16000 (Anthropic requires 16000). temperature, top_p, stop and extended thinking are unsupported. OpenAI/Anthropic SDK text, streaming and tool workflows have been verified; these agent guides are configuration references, not certification of every client feature.
Troubleshooting
Missing models: check configuration version, provider ID and the models list. 401: check your key; 402: insufficient available balance; 400: unsupported parameters or images; 429: reduce concurrency; 503: maintenance or unavailable service. Interrupted requests with unknown usage may retain a reservation; check usage before retrying.
Good to know
What is a token? Is it cryptocurrency?
Tokens measure the text a model processes. They are not cryptocurrency. Balances are for personal use here and cannot be transferred, traded, or cashed out.
Can I watch ads and call the API now?
Model APIs are connected and require available tokens. Ads are unavailable; new accounts do not receive automatic test credit.
What is the context limit?
The upstream enforces a 100K context limit, including history, system prompts and tool definitions. Oversized contexts return errors and are never silently truncated. Request bodies are limited to 2 MiB.
What are the concurrency and reward limits?
Up to 20 concurrent requests per account, with a temporary site-wide limit of 40. Each request reserves 116,000 tokens; actual input plus output is charged and the remainder released. Ad reward rules are pending.
Are all OpenAI parameters supported?
Supports model lists, text, streaming and function tools in Chat Completions and Messages. Sampling, images, extended thinking and forced tool choice are unavailable. Output ceiling is fixed at 16,000; Anthropic max_tokens must be 16000. Unsupported parameters return errors.