agentbox-inference

Compare original and translation side by side

🇺🇸

Original

English

🇨🇳

Translation

Chinese

LLM Inference

LLM推理

Paid OpenAI-compatible chat completions API at

https://inference.x402.agentbox.fyi

. Costs $0.001-$0.003 USDC per call via x402 on Solana. Use the

x_payment

tool for all requests.

付费的兼容OpenAI的聊天补全API，地址为

https://inference.x402.agentbox.fyi

。通过Solana上的x402进行支付，每次调用费用为$0.001-$0.003 USDC。所有请求均需使用

x_payment

工具。

Endpoint

端点

Chat Completions

聊天补全

Generate a chat completion from a supported model.

x_payment({
  "url": "https://inference.x402.agentbox.fyi/v1/chat/completions",
  "method": "POST",
  "body": "{\"model\": \"moonshotai/kimi-k2.5\", \"messages\": [{\"role\": \"user\", \"content\": \"Explain x402 in one sentence\"}]}"
})

Body Parameters:

Param	Type	Required	Description
model	string	yes	Model ID (see table below)
messages	array	yes	Array of `{role, content}` objects
max_tokens	integer	no	Maximum tokens to generate
temperature	number	no	Sampling temperature (0-2)
top_p	number	no	Nucleus sampling (0-1)

Message roles:

system

user

assistant

从支持的模型生成聊天补全结果。

x_payment({
  "url": "https://inference.x402.agentbox.fyi/v1/chat/completions",
  "method": "POST",
  "body": "{\"model\": \"moonshotai/kimi-k2.5\", \"messages\": [{\"role\": \"user\", \"content\": \"Explain x402 in one sentence\"}]}"
})

请求体参数：

参数	类型	是否必填	描述
model	字符串	是	模型ID（见下表）
messages	数组	是	包含 `{role, content}` 对象的数组
max_tokens	整数	否	生成的最大tokens数量
temperature	数字	否	采样温度（0-2）
top_p	数字	否	核采样（0-1）

消息角色：

system

、

user

、

assistant

Models & Pricing

模型与定价

Model	Cost/call	Best for
`moonshotai/kimi-k2.5`	$0.003	High-quality output, large context (262K)
`minimax/minimax-m2.5`	$0.002	Balanced quality/cost

模型	每次调用费用	适用场景
`moonshotai/kimi-k2.5`	$0.003	高质量输出、大上下文（262K）
`minimax/minimax-m2.5`	$0.002	平衡质量与成本

Usage Patterns

使用模式

Simple question

简单问题

x_payment({
  "url": "https://inference.x402.agentbox.fyi/v1/chat/completions",
  "method": "POST",
  "body": "{\"model\": \"moonshotai/kimi-k2.5\", \"messages\": [{\"role\": \"user\", \"content\": \"What is the x402 protocol?\"}]}"
})

x_payment({
  "url": "https://inference.x402.agentbox.fyi/v1/chat/completions",
  "method": "POST",
  "body": "{\"model\": \"moonshotai/kimi-k2.5\", \"messages\": [{\"role\": \"user\", \"content\": \"What is the x402 protocol?\"}]}"
})

With system prompt and parameters

带系统提示词与参数

x_payment({
  "url": "https://inference.x402.agentbox.fyi/v1/chat/completions",
  "method": "POST",
  "body": "{\"model\": \"moonshotai/kimi-k2.5\", \"messages\": [{\"role\": \"system\", \"content\": \"You are a concise technical writer.\"}, {\"role\": \"user\", \"content\": \"Write a summary of Solana's transaction model\"}], \"max_tokens\": 500, \"temperature\": 0.7}"
})

x_payment({
  "url": "https://inference.x402.agentbox.fyi/v1/chat/completions",
  "method": "POST",
  "body": "{\"model\": \"moonshotai/kimi-k2.5\", \"messages\": [{\"role\": \"system\", \"content\": \"You are a concise technical writer.\"}, {\"role\": \"user\", \"content\": \"Write a summary of Solana's transaction model\"}], \"max_tokens\": 500, \"temperature\": 0.7}"
})

Response Format

响应格式

Standard OpenAI chat completion response:

json

{
  "id": "gen-...",
  "object": "chat.completion",
  "model": "moonshotai/kimi-k2.5",
  "choices": [{
    "index": 0,
    "message": { "role": "assistant", "content": "..." },
    "finish_reason": "stop"
  }],
  "usage": {
    "prompt_tokens": 12,
    "completion_tokens": 42,
    "total_tokens": 54
  }
}

标准的OpenAI聊天补全响应：

json

{
  "id": "gen-...",
  "object": "chat.completion",
  "model": "moonshotai/kimi-k2.5",
  "choices": [{
    "index": 0,
    "message": { "role": "assistant", "content": "..." },
    "finish_reason": "stop"
  }],
  "usage": {
    "prompt_tokens": 12,
    "completion_tokens": 42,
    "total_tokens": 54
  }
}

Errors

错误说明

HTTP	Meaning
400	Invalid request (check model name and messages format)
402	Payment required (handled automatically by x_payment)
502	Upstream provider error

HTTP状态码	含义
400	请求无效（检查模型名称和消息格式）
402	需要支付（由x_payment自动处理）
502	上游服务商错误

Cost

费用说明

Flat rate per model per call. Price is determined by the

model

field in the request body. Each call is independent - no sessions or state.

按模型按次收取固定费用。价格由请求体中的

model

字段决定。每次调用相互独立——无会话或状态留存。