{ "displayName": "ZhipuAI", "baseUrl": "https://open.bigmodel.cn/api/paas/v4", "apiKeyTemplate": "xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx.xxxxxxxxxxxxxxxx", "models": [ { "id": "glm-5.2", "name": "GLM-5.2 (CodingPlan)", "sdkMode": "anthropic", "baseUrl": "https://open.bigmodel.cn/api/anthropic", "tooltip": "GLM-5.2 是面向长任务时代的旗舰模型。支持真正可用的 1M 上下文,实测可承载项目级工程上下文,长程任务执行更稳定、工程规范遵循更可靠,开发场景成功率进一步提升。一次任务即可完成“从需求到多端可部署产物”的完整开发链路。", "contextSize": [1000000, 512000, 400000, 256000, 192000], "maxInputTokens": 936000, "maxOutputTokens": 64000, "reasoningEffort": ["high", "max", "none"], "capabilities": { "toolCalling": true, "imageInput": false, "editTools": true }, "tokenPricing": { "USD": [1.4, 4.4, 0.26], "RMB": [8, 28, 2] } }, { "id": "glm-5v-turbo", "name": "GLM-5V-Turbo (CodingPlan)", "sdkMode": "anthropic", "baseUrl": "https://open.bigmodel.cn/api/anthropic", "tooltip": "GLM-5V-Turbo 是智谱首个多模态 Coding 基座模型,面向视觉编程任务打造。能够原生处理图片、视频、文本等多模态输入,同时擅长长程规划、复杂编程和动作执行;深度适配 Agent 工作流,能够与 Claude Code、OpenClaw 等 Agent 深度协同,完成”看懂环境→规划动作→执行任务”的完整闭环。", "maxInputTokens": 168000, "maxOutputTokens": 32000, "thinking": ["enabled", "disabled"], "capabilities": { "toolCalling": true, "imageInput": true, "editTools": true }, "tokenPricing": { "USD": [1.2, 4, 0.24], "RMB": [5, 22, 1.2] } }, { "id": "glm-5-turbo", "name": "GLM-5-Turbo (CodingPlan)", "sdkMode": "anthropic", "baseUrl": "https://open.bigmodel.cn/api/anthropic", "tooltip": "GLM-5-Turbo 是面向 OpenClaw 龙虾场景深度优化的基座模型。 其从训练阶段就针对龙虾任务的核心需求进行专项优化,增强如工具调用、指令遵循、定时与持续性任务、长链路执行等核心能力,使其在复杂、动态、长链路的任务中也真正具备可执行性。", "maxInputTokens": 168000, "maxOutputTokens": 32000, "thinking": ["enabled", "disabled"], "capabilities": { "toolCalling": true, "imageInput": false, "editTools": true }, "tokenPricing": { "pricing": { "USD": [1.2, 4, 0.24], "RMB": [5, 22, 1.2] }, "tiers": [{ "cron": "* 14-17 * * *", "pricing": 3 }] } }, { "id": "glm-4.7", "name": "GLM-4.7 (CodingPlan)", "sdkMode": "anthropic", "baseUrl": "https://open.bigmodel.cn/api/anthropic", "tooltip": "GLM-4.7 是智谱最新旗舰模型,GLM-4.7 面向 Agentic Coding 场景强化了编码能力、长程任务规划与工具协同,并在多个公开基准的当期榜单中取得开源模型中的领先表现。通用能力提升,回复更简洁自然,写作更具沉浸感。在执行复杂智能体任务,在工具调用时指令遵循更强,Artifacts 与 Agentic Coding 的前端美感和长程任务完成效率进一步提升。", "maxInputTokens": 168000, "maxOutputTokens": 32000, "thinking": ["enabled", "disabled"], "capabilities": { "toolCalling": true, "imageInput": false, "editTools": true }, "tokenPricing": { "USD": [0.6, 2.2, 0.11], "RMB": [2, 8, 0.4] } }, { "id": "glm-4.6", "name": "GLM-4.6 (CodingPlan)", "sdkMode": "anthropic", "baseUrl": "https://open.bigmodel.cn/api/anthropic", "tooltip": "GLM-4.6 是智谱的语言模型,其总参数量 355B,激活参数 32B。GLM-4.6 所有核心能力上均完成了对 GLM-4.5 的超越", "maxInputTokens": 168000, "maxOutputTokens": 32000, "thinking": ["disabled", "enabled"], "capabilities": { "toolCalling": true, "imageInput": false, "editTools": true }, "tokenPricing": { "USD": [0.6, 2.2, 0.11], "RMB": [2, 8, 0.4] } }, { "id": "glm-4.6v", "model": "glm-4.6v", "name": "GLM-4.6V (CodingPlan)", "baseUrl": "https://open.bigmodel.cn/api/coding/paas/v4", "tooltip": "ZHIPU GLM-4.6V - 智谱最新视觉推理模型,视觉理解精度达同规模 SOTA,全面支持工具调用,支持 128k 超长上下文,并针对 Coding 场景进行了专项优化。", "maxInputTokens": 112000, "maxOutputTokens": 16000, "thinking": ["disabled", "enabled"], "capabilities": { "toolCalling": true, "imageInput": true, "editTools": true }, "tokenPricing": { "USD": [0.3, 0.9, 0.05], "RMB": [1, 3, 0.2] } }, { "id": "glm-5.2-billing", "model": "glm-5.2", "name": "GLM-5.2 (PayGo)", "tooltip": "GLM-5.2 是面向长任务时代的旗舰模型。支持真正可用的 1M 上下文,实测可承载项目级工程上下文,长程任务执行更稳定、工程规范遵循更可靠,开发场景成功率进一步提升。一次任务即可完成“从需求到多端可部署产物”的完整开发链路。", "contextSize": [1000000, 512000, 400000, 256000, 192000], "maxInputTokens": 936000, "maxOutputTokens": 64000, "reasoningEffort": ["high", "max", "none"], "thinkingFormat": "object-none", "capabilities": { "toolCalling": true, "imageInput": false }, "tokenPricing": { "USD": [1.4, 4.4, 0.26], "RMB": [8, 28, 2] } }, { "id": "glm-5.1-billing", "model": "glm-5.1", "name": "GLM-5.1 (PayGo)", "tooltip": "GLM-5.1 是智谱最新旗舰模型,代码能力大大增强,长程任务显著提升,能够在单次任务中持续、自主地工作长达 8 小时,完成从规划、执行到迭代优化的完整闭环,交付工程级成果。", "maxInputTokens": 168000, "maxOutputTokens": 32000, "thinking": ["enabled", "disabled"], "capabilities": { "toolCalling": true, "imageInput": false }, "tokenPricing": { "pricing": { "USD": [1.4, 4.4, 0.26], "RMB": [6, 24, 1.3] }, "tiers": [{ "contextSizeMin": 32000, "pricing": { "USD": [1.4, 4.4, 0.26], "RMB": [8, 28, 2] } }] } }, { "id": "glm-5v-turbo-billing", "model": "glm-5v-turbo", "name": "GLM-5V-Turbo (PayGo)", "tooltip": "GLM-5V-Turbo 是智谱首个多模态 Coding 基座模型,面向视觉编程任务打造。能够原生处理图片、视频、文本等多模态输入,同时擅长长程规划、复杂编程和动作执行;深度适配 Agent 工作流,能够与 Claude Code、OpenClaw 等 Agent 深度协同,完成”看懂环境→规划动作→执行任务”的完整闭环。", "maxInputTokens": 168000, "maxOutputTokens": 32000, "thinking": ["enabled", "disabled"], "capabilities": { "toolCalling": true, "imageInput": true }, "tokenPricing": { "pricing": { "USD": [1.2, 4, 0.24], "RMB": [5, 22, 1.2] }, "tiers": [{ "contextSizeMin": 32000, "pricing": { "USD": [1.2, 4, 0.24], "RMB": [7, 26, 1.8] } }] } }, { "id": "glm-5-turbo-billing", "model": "glm-5-turbo", "name": "GLM-5-Turbo (PayGo)", "tooltip": "GLM-5-Turbo 是面向 OpenClaw 龙虾场景深度优化的基座模型。 其从训练阶段就针对龙虾任务的核心需求进行专项优化,增强如工具调用、指令遵循、定时与持续性任务、长链路执行等核心能力,使其在复杂、动态、长链路的任务中也真正具备可执行性。", "maxInputTokens": 168000, "maxOutputTokens": 32000, "thinking": ["enabled", "disabled"], "capabilities": { "toolCalling": true, "imageInput": false }, "tokenPricing": { "pricing": { "USD": [1.2, 4, 0.24], "RMB": [5, 22, 1.2] }, "tiers": [{ "contextSizeMin": 32000, "pricing": { "USD": [1.2, 4, 0.24], "RMB": [7, 26, 1.8] } }] } }, { "id": "glm-5-billing", "model": "glm-5", "name": "GLM-5 (PayGo)", "tooltip": "GLM-5 是智谱新一代的旗舰基座模型,面向 Agentic Engineering 打造,能够在复杂系统工程与长程 Agent 任务中提供可靠生产力。在 Coding 与 Agent 能力上,GLM-5 取得开源 SOTA 表现,在真实编程场景的使用体感逼近 Claude Opus 4.5,擅长复杂系统工程与长程 Agent 任务,是通用 Agent 助手的理想基座。", "maxInputTokens": 168000, "maxOutputTokens": 32000, "thinking": ["enabled", "disabled"], "capabilities": { "toolCalling": true, "imageInput": false }, "tokenPricing": { "pricing": { "USD": [1, 3.2, 0.2], "RMB": [4, 18, 1] }, "tiers": [{ "contextSizeMin": 32000, "pricing": { "USD": [1, 3.2, 0.2], "RMB": [6, 22, 1.5] } }] } }, { "id": "glm-4.7-billing", "model": "glm-4.7", "name": "GLM-4.7 (PayGo)", "tooltip": "GLM-4.7 是智谱最新旗舰模型,GLM-4.7 面向 Agentic Coding 场景强化了编码能力、长程任务规划与工具协同,并在多个公开基准的当期榜单中取得开源模型中的领先表现。通用能力提升,回复更简洁自然,写作更具沉浸感。在执行复杂智能体任务,在工具调用时指令遵循更强,Artifacts 与 Agentic Coding 的前端美感和长程任务完成效率进一步提升。", "maxInputTokens": 168000, "maxOutputTokens": 32000, "thinking": ["enabled", "disabled"], "capabilities": { "toolCalling": true, "imageInput": false }, "tokenPricing": { "pricing": { "USD": [0.6, 2.2, 0.11], "RMB": [2, 8, 0.4] }, "tiers": [{ "contextSizeMin": 32000, "pricing": { "USD": [0.6, 2.2, 0.11], "RMB": [4, 16, 0.8] } }] } }, { "id": "glm-4.6-billing", "model": "glm-4.6", "name": "GLM-4.6 (PayGo)", "tooltip": "GLM-4.6 是智谱的语言模型,其总参数量 355B,激活参数 32B。GLM-4.6 所有核心能力上均完成了对 GLM-4.5 的超越", "maxInputTokens": 168000, "maxOutputTokens": 32000, "thinking": ["disabled", "enabled"], "capabilities": { "toolCalling": true, "imageInput": false }, "tokenPricing": { "pricing": { "USD": [0.6, 2.2, 0.11], "RMB": [2, 8, 0.4] }, "tiers": [{ "contextSizeMin": 32000, "pricing": { "USD": [0.6, 2.2, 0.11], "RMB": [4, 16, 0.8] } }] } }, { "id": "glm-4.6v-billing", "model": "glm-4.6v", "name": "GLM-4.6V (PayGo)", "tooltip": "GLM-4.6V - 智谱最新视觉推理模型,视觉理解精度达同规模 SOTA,全面支持工具调用,支持 128k 超长上下文,并针对 Coding 场景进行了专项优化。", "maxInputTokens": 112000, "maxOutputTokens": 16000, "thinking": ["disabled", "enabled"], "capabilities": { "toolCalling": true, "imageInput": true }, "tokenPricing": { "USD": [0.3, 0.9, 0.05], "RMB": [1, 3, 0.2] } }, { "id": "glm-4.7-flash", "name": "GLM-4.7-Flash (Free)", "tooltip": "GLM-4.7-Flash 作为 30B 级 SOTA 模型,提供了一个兼顾性能与效率的新选择。面向 Agentic Coding 场景强化了编码能力、长程任务规划与工具协同,并在多个公开基准的当期榜单中取得同尺寸开源模型中的领先表现。在执行复杂智能体任务,在工具调用时指令遵循更强,Artifacts 与 Agentic Coding 的前端美感和长程任务完成效率进一步提升。", "maxInputTokens": 168000, "maxOutputTokens": 32000, "thinking": ["enabled", "disabled"], "capabilities": { "toolCalling": true, "imageInput": false } }, { "id": "glm-4.7-flashx", "name": "GLM-4.7-FlashX (PayGo)", "tooltip": "GLM-4.7-Flash 作为 30B 级 SOTA 模型,提供了一个兼顾性能与效率的新选择。面向 Agentic Coding 场景强化了编码能力、长程任务规划与工具协同,并在多个公开基准的当期榜单中取得同尺寸开源模型中的领先表现。在执行复杂智能体任务,在工具调用时指令遵循更强,Artifacts 与 Agentic Coding 的前端美感和长程任务完成效率进一步提升。", "maxInputTokens": 168000, "maxOutputTokens": 32000, "thinking": ["enabled", "disabled"], "capabilities": { "toolCalling": true, "imageInput": false }, "tokenPricing": { "USD": [0.07, 0.4, 0.01], "RMB": [0.5, 3, 0.1] } }, { "id": "glm-4.6v-flash", "name": "GLM-4.6V-Flash (Free)", "tooltip": "GLM-4.6V-Flash 是 GLM-4.6V 的免费版本,是 GLM 系列在多模态方向上的一次重要迭代,支持开启或关闭思考模式。它将训练时上下文窗口提升到128k tokens,在 视觉理解精度上达到同参数规模 SOTA,并首次在模型架构中将 Function Call(工具调用)能力原生融入视觉模型,打通从「视觉感知」到「可执行行动(Action)」的链路,为真实业务场景中的多模态 Agent 提供统一的技术底座。", "maxInputTokens": 112000, "maxOutputTokens": 16000, "thinking": ["disabled", "enabled"], "capabilities": { "toolCalling": true, "imageInput": true } } ] }