Grok 4.20 (Non-Reasoning)

高性能非推理模型，侧重速度、智能体工具调用、低幻觉率和严格遵循提示词。

xAIgrok-4.20-0309-non-reasoning2026-03-10报告数据错误

规格

上下文: 1M
最大输出: —
输入价: $1.25/ 百万 Token
输出价: $2.5/ 百万 Token
发布日期: 2026-03-10

上下文

上下文: 1M
最大输入: —
最大输出: —

价格 · USD / 百万 Token

输入: $1.25
输出: $2.5
缓存读取: $0.2
缓存写入: —

输入输出

输入: 文本图像
输出: 文本

API

接口类型: chatresponsesbatch

思考

支持思考: 不支持

源信息

状态: 可用
发布日期: 2026-03-10
知识截止: —

能力

已确认: toolsstructured_outputcachingbatch

调用方式

xaimodel = grok-4.20-0309-non-reasoning

chat

POSThttps://api.x.ai/v1/chat/completions

responses

POSThttps://api.x.ai/v1/responses

JSON

grok-4.20-0309-non-reasoning.json

{
  "id": "grok-4.20-0309-non-reasoning",
  "object": "model",
  "created": 1773100800,
  "owned_by": "xai",
  "name": "Grok 4.20 (Non-Reasoning)",
  "api": {
    "types": [
      "chat",
      "responses",
      "batch"
    ]
  },
  "limits": {
    "context": 1000000,
    "input": null,
    "output": null
  },
  "modalities": {
    "input": [
      "text",
      "image"
    ],
    "output": [
      "text"
    ]
  },
  "reasoning": {
    "supported": false,
    "efforts": []
  },
  "pricing": {
    "currency": "USD",
    "unit": "1M_tokens",
    "input": 1.25,
    "output": 2.5,
    "cache_read": 0.2,
    "cache_write": null
  },
  "features": [
    "tools",
    "structured_output",
    "caching",
    "batch"
  ],
  "info": {
    "status": "active",
    "release_date": "2026-03-10",
    "knowledge_cutoff": null,
    "description": "A high-performance, non-reasoning model focused on speed, agentic tool calling, low hallucination rates, and strict prompt adherence.",
    "docs": "https://docs.x.ai/developers/models/grok-4.20-0309-non-reasoning",
    "verified_at": "2026-10-11"
  }
}

官方来源