Grok's flagship reasoning model, running an internal chain of thought before responding. Strong performance on coding and STEM tasks, with support for text, image, and file inputs. Suited for technical problem-solving and long-document analysis.
| 规格 | 价格 |
|---|---|
| 输入 | $2/ M tokens |
| 输出 | $6/ M tokens |
| 缓存读 | $0.5/ M tokens |
| 缓存写 | -- |
可直接粘贴运行下方示例代码。Gate.AI API 与 OpenAI 兼容，多数 SDK 只需替换 base URL；切换模型时，只需修改模型 ID。
Sign in to the Gate.AI console and create a new key under "API Keys". Keep it safe and send it via the Authorization header on every request.
前往控制台创建密钥Send a POST request to the Chat Completions endpoint with a model ID and an array of messages to get a reply.
curl --location "https://api.gate.ai/openai/v1/chat/completions" \
--header "Authorization: Bearer {API_KEY}" \
--header "Content-Type: application/json" \
--data '{
"model": "grok-4.5",
"messages": [
{"role": "user", "content": "Hello"}
]
}'By default a full JSON response is returned. Add "stream": true to the request body to receive incremental chunks over SSE.
curl --no-buffer --location "https://api.gate.ai/openai/v1/chat/completions" \
--header "Authorization: Bearer {API_KEY}" \
--header "Content-Type: application/json" \
--data '{
"model": "grok-4.5",
"messages": [
{"role": "user", "content": "Hello"}
],
"stream": true
}'基础 URL、请求方法和必需请求头均按当前模型的协议模板渲染。
Grok 4.6Grok's agent-oriented release, built for long-running multi-step work across a codebase. Accepts text, image, and file inputs. Suited for research and interactive or visual output.
Grok 4.3Grok 4.3 is a reasoning model from xAI that supports text and image inputs with text output. It is suited for agentic workflows, instruction following, long-document analysis, and deep research. The model supports a 1M token context window. Requests above 200K total tokens are billed at a higher rate.
Grok 4.20Reasoning-oriented Grok variant built for multi-step problem solving and agentic tool use, with strict prompt adherence and low hallucination.
Grok 4.20 Multi-AgentMulti-agent variant that dispatches sub-agents in parallel for deep research, multi-step reasoning, and tool orchestration.
Grok 4.20 V2 ReasoningReasoning-oriented Grok variant built for multi-step problem solving and agentic tool use, with strict prompt adherence and low hallucination.