{
"id": "chatcmpl-b6ea00c78f686daa",
"object": "chat.completion",
"created": 1791939594,
"model": "qwen3-coder-30b",
"choices": [
{
"index": 0,
"message": {
"role": "assistant",
"content": "READY",
"refusal": null,
"annotations": null,
"audio": null,
"function_call": null,
"reasoning": null
},
"logprobs": null,
"finish_reason": "stop",
"stop_reason": null,
"token_ids": null,
"routed_experts": null
}
],
"service_tier": null,
"system_fingerprint": "vllm-0.31.0-6fef36b3",
"usage": {
"prompt_tokens": 10,
"total_tokens": 12,
"completion_tokens": 2,
"prompt_tokens_details": null,
"completion_tokens_details": null
},
"prompt_logprobs": null,
"prompt_token_ids": null,
"prompt_text": null,
"kv_transfer_params": null,
"ec_transfer_params": null,
"metrics": null,
"credits": {
"deducted": 6,
"tokens": 12,
"remaining": 5019186
}
}
curl --location --request POST 'https://zylalabs.com/api/14252/openai-compatible+llm+chat+completions+api/35372/create+a+chat+completion' --header 'Authorization: Bearer YOUR_API_KEY'
--data-raw '{"model":"qwen3-coder-30b","messages":[{"role":"user","content":"Say READY"}],"max_tokens":16}'
{
"object": "list",
"data": [
{
"id": "synap-v1",
"object": "model",
"owned_by": "synap",
"nodes": 0,
"routing": "auto — routes inside the models we run on our own GPUs and picks the cheapest that can do the job (tools/code/reasoning), billed at the price of whichever model actually ran"
},
{
"id": "synap-v1-pro",
"object": "model",
"owned_by": "synap",
"nodes": 0,
"routing": "auto — adds the external models we offer to our own: the strongest model per job class (long context, reasoning, multilingual) rather than the cheapest"
},
{
"id": "team-planner",
"object": "model",
"owned_by": "linkrra-team-agent",
"served_by": "deepseek/deepseek-v4-pro",
"nodes": 0,
"pricing": {
"prompt_per_million": 2.2788,
"completion_per_million": 4.0512,
"currency": "USD",
"basis": "brokered"
},
"capability": "Reasoning, deep",
"tier": "premium",
"context_note": "DeepSeek's flagship, for problems that are genuinely hard.",
"secure": true,
"data_policy": "zero-retention (verified live via tools/verify-zdr.js)"
},
{
"id": "team-worker",
"object": "model",
"owned_by": "linkrra-team-agent",
"served_by": "qwen3-coder-30b",
"nodes": 0,
"pricing": {
"prompt_per_million": 0.0675,
"completion_per_million": 0.27,
"currency": "USD",
"basis": "brokered"
},
"capability": "Coding, everyday",
"tier": "standard",
"context_note": "Runs on our own GPUs. Your prompt never leaves our infrastructure.",
"secure": true,
"data_policy": "zero-retention by construction — served on Linkrra-operated fleet hardware; the request is never handed to a third-party inference provider"
},
{
"id": "gpt-oss-20b",
"object": "model",
"owned_by": "pearlpool-fleet",
"nodes": 1,
"pricing": {
"prompt_per_million": 0.017,
"completion_per_million": 0.085,
"currency": "USD",
"basis": "engine"
},
"capability": "Fast + cheap general model",
"tier": "standard",
"context_note": "Runs on our own GPUs. Your prompt never leaves our infrastructure.",
"secure": true,
"data_policy": "zero-retention by construction — served on Linkrra-operated fleet hardware; the request is never handed to a third-party inference provider",
"schema_version": "2.4",
"name": "OpenAI: gpt-oss-20b",
"input_modalities": [
{
"type": "text",
"supported_inputs": {
"max_context_length": {
"value": 32768,
"unit": "token"
}
},
"pricing": [
{
"type": "prompt",
"unit": "token",
"cost_usd": "0.000000017"
}
]
}
],
"output_modalities": [
{
"type": "text",
"max_length": {
"value": 32768,
"unit": "token"
},
"streaming": true,
"supported_parameters": {
"temperature": {
"type": "range",
"min": 0,
"max": 2
},
"top_p": {
"type": "range",
"min": 0,
"max": 1
},
"max_tokens": {
"type": "integer",
"min": 1,
"max": 32768,
"unit": "token"
},
"stop": {
"type": "array",
"max_items": 4
}
},
"pricing": [
{
"type": "completion",
"unit": "token",
"cost_usd": "0.000000085"
}
]
}
],
"is_ready": true,
"is_free": false,
"created": 1791849600,
"datacenters": [
{
"country_code": "KR"
}
]
},
{
"id": "qwen3-coder-30b",
"object": "model",
"owned_by": "pearlpool-fleet",
"nodes": 2,
"pricing": {
"prompt_per_million": 0.0675,
"completion_per_million": 0.27,
"currency": "USD",
"basis": "engine"
},
"capability": "Coding, everyday",
"tier": "standard",
"context_note": "Runs on our own GPUs. Your prompt never leaves our infrastructure.",
"secure": true,
"data_policy": "zero-retention by construction — served on Linkrra-operated fleet hardware; the request is never handed to a third-party inference provider",
"schema_version": "2.4",
"name": "Qwen: Qwen3 Coder 30B A3B Instruct",
"input_modalities": [
{
"type": "text",
"supported_inputs": {
"max_context_length": {
"value": 32768,
"unit": "token"
}
},
"pricing": [
{
"type": "prompt",
"unit": "token",
"cost_usd": "0.0000000675"
}
]
}
],
"output_modalities": [
{
"type": "text",
"max_length": {
"value": 32768,
"unit": "token"
},
"streaming": true,
"supported_parameters": {
"temperature": {
"type": "range",
"min": 0,
"max": 2
},
"top_p": {
"type": "range",
"min": 0,
"max": 1
},
"max_tokens": {
"type": "integer",
"min": 1,
"max": 32768,
"unit": "token"
},
"stop": {
"type": "array",
"max_items": 4
}
},
"pricing": [
{
"type": "completion",
"unit": "token",
"cost_usd": "0.00000027"
}
]
}
],
"is_ready": true,
"is_free": false,
"created": 1790985600,
"datacenters": [
{
"country_code": "KR"
}
]
},
{
"id": "glm-4.7-flash",
"object": "model",
"owned_by": "pearlpool-fleet",
"nodes": 1,
"pricing": {
"prompt_per_million": 0.055,
"completion_per_million": 0.38,
"currency": "USD",
"basis": "engine"
},
"capability": "Fast general model",
"tier": "standard",
"context_note": "Runs on our own GPUs. Your prompt never leaves our infrastructure.",
"secure": true,
"data_policy": "zero-retention by construction — served on Linkrra-operated fleet hardware; the request is never handed to a third-party inference provider",
"schema_version": "2.4",
"name": "Z.AI: GLM 4.7 Flash",
"input_modalities": [
{
"type": "text",
"supported_inputs": {
"max_context_length": {
"value": 6144,
"unit": "token"
}
},
"pricing": [
{
"type": "prompt",
"unit": "token",
"cost_usd": "0.000000055"
}
]
}
],
"output_modalities": [
{
"type": "text",
"max_length": {
"value": 6144,
"unit": "token"
},
"streaming": true
}
]
}
],
"_note": "Response truncated for documentation purposes"
}
curl --location --request GET 'https://zylalabs.com/api/14252/openai-compatible+llm+chat+completions+api/35373/list+available+models' --header 'Authorization: Bearer YOUR_API_KEY'
注册后,每个开发者都会被分配一个个人 API 访问密钥,这是一个唯一的字母和数字组合,用于访问我们的 API 端点。要使用 与OpenAI兼容的LLM聊天完成 API 进行身份验证,只需在 Authorization 标头中包含您的 bearer token。
| 标头 | 描述 |
|---|---|
授权
|
必需
应为 Bearer access_key. 订阅后,请查看上方的"您的 API 访问密钥"。
|
POST 创建聊天完成端点返回一个聊天完成对象,包括使用的模型、生成的消息内容和使用统计信息 GET 列出可用模型端点返回一个模型列表,详细说明它们的ID、上下文长度和路由信息
聊天完成响应中的关键字段包括“id”“model”“choices”(带有消息内容)和“usage”(详细的令牌计数)在模型列表响应中重要字段包括“id”“owned_by”和“routing”
聊天完成响应的结构是一个带有嵌套字段的JSON对象,其中包含选择和使用情况。模型列表响应是一个模型对象数组,每个对象包含特定属性,如ID和路由细节
POST 创建聊天完成端点接受参数,如“模型”、“消息”和“流”。用户可以通过指定模型 ID 和提供一系列消息来定制请求以供上下文。
典型的用例包括构建聊天机器人 生成对话代理 和增强客户支持系统 该API可用于创建需要自然语言理解和生成的交互式应用程序
用户可以从“choices”数组中提取“content”字段以显示生成的消息 “usage”字段有助于跟踪令牌消耗 这对优化请求和管理成本很有用
通过使用基于多样化数据集训练的先进模型来维护数据的准确性 该API在Linkrra的硬件上运行 确保生成响应时的一致性能和可靠性
如果响应包含空结果或部分结果,用户应检查“choices”数组以获取可用的消息 实施错误处理可以帮助管理意外输出并确保顺畅的用户体验