跳到主要内容

OpenAILLM

GPT-6 Astra API

模型 IDopenai/gpt-6-astra-saver

OpenAI's GPT-6 Astra on the OpenAI-compatible chat completions API, with a 1,050,000 token context, vision, tools and reasoning, plus a lower-priced saver channel.

  • 1.05M context
  • 工具
  • 视觉
  • 推理
  • 省钱通道

在 Playground 中运行 GPT-6 Astra

Send a chat request with your account and see the exact token charge. Code needs an API key.

参数

模型
model
openai/gpt-6-astra-saver
messages[].content
messages[0].content
max_tokens
temperature

请求预览

{"model": "openai/gpt-6-astra-saver","messages": [],"stream": true,"max_tokens": 1024}

网页 Playground 使用你的登录会话发送此请求体。在代码中请使用 API Key 发送相同内容。

结果

发送一条消息,用你的账号运行 GPT-6 Astra。

收到回复后,这里会显示用量和扣费。

GPT-6 Astra 价格

美元 / 百万 token。按每次请求的提示词长度选择档位。

正在加载价格…
  1. Reserve请求被接受时,Hermes 会按估算的提示词长度加上 max_tokens(未设置时为 8,192)预留额度。设置 max_tokens 可以减少预留。
  2. Settle响应结束后,按模型返回的实际用量结算:提示词 token 按输入价、命中缓存的 token 按缓存读取价、输出 token 按输出价,预留的剩余部分立即释放。
  3. Failures模型拒绝请求时不收费。流式输出在已生成内容后中断时,会按已生成文本估算并扣费。

对比 GPT-6 Astra 各版本

The saver channel is the same model on a lower-priority upstream route, with its own model ID and price.

版本模型 ID上下文窗口能力
Standardopenai/gpt-6-astra1,050,000 token工具, 视觉, 推理
省钱版openai/gpt-6-astra-saver1,050,000 token工具, 视觉, 推理

GPT-6 Astra 能做什么

OpenAI positions GPT-6 Astra for demanding reasoning, coding, research and document creation. Its official Responses API lists additional hosted tools; those manufacturer capabilities should not be assumed to be exposed by the Hermes chat-completions route.

GPT-6 Astra is served through the OpenAI-compatible endpoint, so an existing OpenAI SDK only needs the Hermes base URL and API key. The catalog lists a context window of 1,050,000 tokens and support for tools, vision and reasoning.

The family has two model IDs. openai/gpt-6-astra is the standard channel; openai/gpt-6-astra-saver is the same model on a lower-priority upstream route with its own price. The manufacturer list price has two tiers split at 272,000 prompt tokens, and the pricing table shows which tier a request falls in.

使用场景

  1. Agents with function tools

    Pass tools in the request; the model returns tool calls for your code to run.

  2. Questions about images

    Send screenshots or photos as image_url parts next to the text.

  3. Long inputs

    Work with long documents or code within the listed context, allowing space for the response and other request tokens.

用代码调用 GPT-6 Astra

任意 OpenAI SDK 都可用:设置 base URL 和你的 Hermes Key。

  1. 1

    创建 API Key

    In the dashboard, with the chat:create scope. API 密钥
  2. 2

    Point your SDK at Hermes

    Base URL https://api.hermes-ai.net/api/v1
  3. 3

    发送对话补全请求

    POST /chat/completions with openai/gpt-6-astra-saver

代码示例

Set HERMES_API_KEY in your environment.

shell
export HERMES_API_KEY=your_api_key
Full chat API reference
curl https://api.hermes-ai.net/api/v1/chat/completions \
  -H "Authorization: Bearer $HERMES_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "openai/gpt-6-astra-saver",
    "messages": [{"role": "user", "content": "Say hello in one sentence."}],
    "max_tokens": 200
  }'

请求

POST https://api.hermes-ai.net/api/v1/chat/completions 带 chat:create 权限的 API Key,或已登录会话。

字段类型说明
modelstringopenai/gpt-6-astra-saver
messagesarrayOpenAI 对话消息:system、user、assistant、tool 角色;text 和 image_url 内容块。
streambooleanSSE 流式输出,最后一个事件包含 usage。
max_tokensinteger输出上限,同时决定预留额度。
tools / tool_choicearray / object仅支持 function 工具。
response_formatobject模型支持时可用 JSON 模式和 JSON schema。
temperature, top_p, stop, seed, reasoning_effort原样透传。

响应

字段说明
choices[].message / deltaOpenAI chat completion 结构。
usage模型返回的 prompt_tokens、completion_tokens 和 prompt_tokens_details.cached_tokens。
hermes.charged_micros精确扣费(微美元,JSON 响应)。流式请求可在 GET /api/v1/chat/requests 查看。
x-hermes-request请求 id 响应头,账单流水中使用同一 id。
error.code文档中的 Hermes 错误码;请按错误码而非消息文本分支。

GPT-6 Astra API 常见问题

Answers built from the model's catalog entry and the Hermes AI API contract.

When should I use the saver channel?

When a lower price matters more than priority. It is the same model on a lower-priority route; send openai/gpt-6-astra-saver as the model to use it.

Why are there two price tiers?

The tier is chosen by the prompt size of each request: prompts up to 272,000 tokens use the first tier, longer prompts the second.

What is the GPT-6 Astra API on Hermes AI?

GPT-6 Astra by OpenAI is served through the OpenAI-compatible chat completions endpoint, https://api.hermes-ai.net/api/v1/chat/completions, with the model ID openai/gpt-6-astra.

What is the context window of GPT-6 Astra?

1,050,000 tokens, shared by the prompt and the output.

Does GPT-6 Astra support vision, tools and JSON output?

Supported: 工具, 视觉 和 推理. Not supported: JSON 和 音频. Streaming works with stream: true.

How is GPT-6 Astra billed?

Per token, from prepaid credits. When a request is accepted, credits are reserved for the estimated prompt plus max_tokens of output (8,192 when you set none); when it finishes, the exact charge is taken from the usage the model reports and the rest is released. A rejected request is not charged. Rates are in the pricing table.

Can I use the OpenAI SDK with GPT-6 Astra?

Yes. Set the base URL to https://api.hermes-ai.net/api/v1 and use your Hermes API key; send openai/gpt-6-astra as the model. Chat completions, streaming, tools and JSON output work unchanged.

What is the GPT-6 Astra saver channel?

The same model on a lower-priority upstream route, with its own model ID (openai/gpt-6-astra-saver) and its own price. Pick it in the variant switch on this page.

What happens when I hit a rate limit?

Too many requests per minute return 429 RATE_LIMITED and too many in flight return 429 CONCURRENCY_LIMITED. If the model provider rate limits, you get 429 UPSTREAM_RATE_LIMITED and nothing is charged. Retry with backoff.

How do I get an API key for GPT-6 Astra?

Create one under API 密钥 in the dashboard with the chat:create scope, then add credits on the Billing page.