此模型已弃用。我们仅继续对默认的 10k 输入 token 工作负载进行性能测试。其他工作负载的结果为历史数据，不再更新。

Z AI 已推出更新版本 GLM-5。我们建议考虑使用该版本。

Z AI

•

GLM-4.7

•

Open weights model

•

Released 2025年12月

GLM-4.7 (Reasoning) 智能、性能与价格分析

立即试用 API 服务商基准测试

智能Updated

22*

Artificial Analysis Intelligence Index

* 估计值

成本

输入 US$0.60输出 US$2.20缓存折扣 25%

无

Intelligence Index 每项任务的成本

冗长度

无

Intelligence Index 输出 token 数

GLM-4.7 (Reasoning) 的智能水平高于平均水平，但价格有些昂贵；比较对象为其他规模相近的开放权重模型。该模型支持文本输入，可输出文本，上下文窗口为 200k 个 token。

GLM-4.7 (Reasoning) 在 Artificial Analysis Intelligence Index 上的得分为 22，在同类模型中高于平均水平（中位数：19）。

GLM-4.7 (Reasoning) 每 100 万输入 token 的价格为 $0.60（有些昂贵，中位数：$0.30），每 100 万输出 token 的价格为 $2.20（有些昂贵，中位数：$1.13）。

GLM-4.7 (Reasoning) 的速度为每秒 67 个 token，低于平均水平（72）。

推理	是此页面展示该模型的推理版本。可能还存在非推理版本。
输入模态	支持：文本
输出模态	支持：文本
上下文窗口	200k 约 300 页 A4 纸（12 号 Arial 字体）
总参数量	357B
活跃参数量	32B 推理期间每个 token 启用的参数数量
许可证	MIT
模型权重	Hugging Face

指标与同类别模型进行比较：

非推理模型 → 仅与其他非推理模型比较
推理模型 → 同时与推理和非推理模型比较
开放权重模型 → 仅与规模类别相同的其他开放权重模型比较：

微型：≤4B 个参数
小型：4B–40B 个参数
中型：40B–150B 个参数
大型：>150B 个参数

专有模型 → 与价格区间相同的专有模型和开放权重模型比较，采用输入/输出价格 3:1 的混合比例：

每 100 万 token <$0.15
每 100 万 token $0.15–$1
每 100 万 token >$1

Updated

智能

Artificial Analysis Intelligence Index · Higher is better

Not publicly available

速度

Output tokens per second · Higher is better

每任务成本

Weighted average cost (USD) per Intelligence Index task · Lower is better

Not publicly available

智能Updated

Artificial Analysis Intelligence Index

Artificial Analysis Intelligence Index v4.3.2 incorporates 10 evaluations: AA-Briefcase v1.1, GDPval-AA v2.1, AutomationBench-AA, Terminal-Bench 4.0, SciCode, Humanity's Last Exam, GDP.pdf, CritPt, AA-Omniscience, AA-LCR v1.1

Not publicly available

Estimate (independent evaluation forthcoming)

Artificial Analysis Intelligence Index v4.3.2 includes: AA-Briefcase v1.1, GDPval-AA v2.1, AutomationBench-AA, Terminal-Bench 4.0, SciCode, Humanity's Last Exam, GDP.pdf, CritPt, AA-Omniscience, AA-LCR v1.1. See Intelligence Index methodology for further details, including a breakdown of each evaluation and how we run them.

Artificial Analysis Intelligence Index：开放权重与专有模型

Not publicly available

Estimate (independent evaluation forthcoming)

Indicates whether the model weights are available. Models are labelled as 'Commercial Use Restricted' if commercial use is limited by conditions, and as 'Non-commercial' if the license prohibits commercial use.

基准测试

Intelligence Evaluations

Intelligence evaluations measured independently by Artificial Analysis · Higher is better

Not publicly available

AA-Briefcase v1.1Updated

Agentic knowledge work, (Elo-500)/2000

GDPval-AA v2.1Updated

Agentic real-world work tasks, (Elo-500)/2000

AutomationBench-AA

Agentic SaaS workflows

Terminal-Bench 4.0

Agentic coding & terminal use

SciCodeUnder review

Coding

Humanity's Last Exam

Reasoning & knowledge

GDP.pdf

Professional document reasoning, All-pass

CritPtUnder review

Physics reasoning

AA-Omniscience Accuracy

Knowledge

AA-Omniscience Non-Hallucination Rate

1 - hallucination rate

AA-LCR v1.1

Long context reasoning

Harvey LAB-AA

Legal agentic work, criterion pass rate

EnterpriseOps-Gym-AA

Agentic business operations

Terminal-Bench-Science 0.1New

Agentic scientific research workflows in a terminal

AA-AnalystAgent

Quantitative analysis on spreadsheets & documents

ITBench-AA

Kubernetes incident root-cause analysis

MMMU-Pro

Visual reasoning

MLCR-AA

Medical long context reasoning

While model intelligence generally translates across use cases, specific evaluations may be more relevant for certain use cases.

AA-Omniscience

AA-Omniscience Index

AA-Omniscience Index (higher is better) measures knowledge reliability and hallucination. It rewards correct answers, penalizes hallucinations, and has no penalty for refusing to answer. Scores range from -100 to 100, where 0 means as many correct as incorrect answers, and negative scores mean more incorrect than correct.

Not publicly available

开放性指数

Artificial Analysis 开放性指数：得分

开放性指数以 0 至 100 的标准化评分衡量模型的开放程度（分数越高，越开放）

Intelligence Index 比较

Intelligence Index 与每项任务的成本

Artificial Analysis Intelligence Index · Weighted average cost (USD) per Artificial Analysis Intelligence Index task

Most attractive quadrant

Pareto line

Weighted average cost per Intelligence Index task. Each evaluation’s cost is calculated from input, cache hit, cache write, reasoning, and answer token prices, divided by task count, and weighted by its Intelligence Index weight.

成本

每项 Intelligence Index 任务的成本

Weighted average cost (USD) per Artificial Analysis Intelligence Index task, segmented by token type. Lower is better

Not publicly available

运行 Artificial Analysis Intelligence Index 的成本

Cost (USD) to run all evaluations in the Artificial Analysis Intelligence Index

Not publicly available

The cost to run the evaluations in the Artificial Analysis Intelligence Index, calculated using the model's input, cache hit, cache write, reasoning, and answer token prices and the number of tokens used across evaluations (excluding repeats).

价格：缓存命中、输入和输出

Price (USD per M Tokens)

Not publicly available

Price per token for cached prompts (previously processed), typically offering a significant discount compared to regular input price, represented as USD per million tokens. The values shown here are the cache hit price; cache write and cache storage are billed separately and vary by provider — see "Cache pricing by provider" for detail.

上下文窗口

Context window: tokens limit · Higher is better

Not publicly available

Larger context windows are relevant to RAG (Retrieval Augmented Generation) LLM workflows which typically involve reasoning and information retrieval of large amounts of data.

Maximum number of combined input & output tokens. Output tokens commonly have a significantly lower limit (varied by model).

速度

按输出速度（每秒 token 数）衡量

输出速度

Output tokens per second · Higher is better

Tokens per second received while the model is generating tokens (ie. after first chunk has been received from the API for models which support streaming).

Figures represent performance of the model's first-party API or the median across providers where a first-party API is not available.

智能指数每项任务耗时

Weighted average decode time (minutes) per task; excludes TTFT and overhead time · Lower is better

The weighted average time (seconds) per Artificial Analysis Intelligence Index task. This is calculated by dividing output tokens per task by output speed, weighted by the relative weights of each benchmark in the Intelligence Index.

延迟

按首 Token 延迟（秒）衡量

延迟: 首个答案 Token 延迟

收到首个回答 token 的秒数 · 包含推理模型的“思考”时间

Time to first answer token received, in seconds, after API request sent. For reasoning models, this includes the 'thinking' time of the model before providing an answer. For models which do not support streaming, this represents time to receive the completion.

端到端响应时间

Seconds to output 500 tokens, calculated based on time to first token, 'thinking' time for reasoning models, and output speed

端到端响应时间

Seconds to output 500 tokens, including reasoning model 'thinking' time · Lower is better

Seconds to receive a 500 token response. Key components:

Input time: Time to receive the first response token
Thinking time (only for reasoning models): Time reasoning models spend outputting tokens to reason prior to providing an answer. Amount of tokens based on the average reasoning tokens across a diverse set of 60 prompts (methodology details).
Answer time: Time to generate 500 output tokens, based on output speed

Figures represent performance of the model's first-party API or the median across providers where a first-party API is not available.

模型规模（仅开放权重模型）

模型规模：总参数量与活跃参数量

Comparison between total model parameters and parameters active during inference (billions)

The total number of trainable weights and biases in the model, expressed in billions. These parameters are learned during training and determine the model's ability to process and generate responses.

The number of parameters actually executed during each inference forward pass, expressed in billions. For Mixture of Experts (MoE) models, a routing mechanism selects a subset of experts per token, resulting in fewer active than total parameters. Dense models use all parameters, so active equals total.

常见问题

关于 GLM-4.7 (Reasoning) 的常见问题

GLM-4.7 (Reasoning) 发布于 2025年12月22日。

GLM-4.7 (Reasoning) 由 Z AI 开发。

GLM-4.7 (Reasoning) 在 Artificial Analysis Intelligence Index 上的估算得分为 22，在其他规模相近的开放权重模型中高于平均水平（中位数：19）。

GLM-4.7 (Reasoning) 以每秒 67.3 个 token 的速度生成输出（基于提供该模型的各服务商中位数），与其他规模相近的开放权重模型相比低于平均水平（中位数：71.9 t/s）。

GLM-4.7 (Reasoning) 的首 Token 延迟（TTFT）为 1.26 秒（基于提供该模型的各服务商中位数），与其他规模相近的开放权重模型相比非常有竞争力（中位数：2.21 秒）。

GLM-4.7 (Reasoning) 每 100 万输入 token 的价格为 $0.60（略高于平均水平，中位数：$0.43），每 100 万输出 token 的价格为 $2.20（略高于平均水平，中位数：$1.68）；基于提供该模型的各服务商中位数。

GLM-4.7 (Reasoning) 每 100 万输入 token 的价格为 $0.60，每 100 万输出 token 的价格为 $2.20（基于提供该模型的各服务商中位数）。按缓存命中/输入/输出为 7:2:1 的比例计算，混合价格为每 100 万 token $0.66。价格可能因服务商而异。比较服务商价格

是，GLM-4.7 (Reasoning) 是推理模型。它会在给出答案之前，通过扩展思考或思维链推理来解决复杂问题。

GLM-4.7 (Reasoning) 支持文本输入。

GLM-4.7 (Reasoning) 支持文本输出。

否，GLM-4.7 (Reasoning) 不支持图像输入，只能处理文本。

否，GLM-4.7 (Reasoning) 不是多模态模型，仅支持文本输入。

GLM-4.7 (Reasoning) 的上下文窗口为 200k 个 token。这决定了模型在单次请求中可以处理多少文本和对话历史。

是，GLM-4.7 (Reasoning) 是开放权重模型，其模型权重已公开，可供下载并自行托管。

GLM-4.7 (Reasoning) 有 357B 参数（启用 32B）。

GLM-4.7 (Reasoning) 是混合专家（MoE）模型，总参数量为 357B，但推理期间仅启用 32B。

GLM-4.7 (Reasoning) 基于 MIT 许可证发布，该许可证允许商业使用。查看许可证

GLM-4.7 (Reasoning) 在 Artificial Analysis Intelligence Index 上的得分为 22。这项综合基准测试评估模型的推理、知识、数学和编程能力。

是，可通过 7 家服务商的 API 使用 GLM-4.7 (Reasoning)。比较 API 服务商

可通过 7 家 API 服务商使用 GLM-4.7 (Reasoning)。比较服务商

GLM-4.7 (Reasoning) 智能、性能与价格分析

模型摘要

智能Updated

速度

成本

冗长度

比较摘要

技术规格

此类别中有 118 个模型

智能

速度

每任务成本

智能Updated

Artificial Analysis Intelligence Index

Artificial Analysis Intelligence Index

Artificial Analysis Intelligence Index：开放权重与专有模型

Artificial Analysis Intelligence Index

Open Weights

基准测试

Intelligence Evaluations

Intelligence Evaluation Relevance

Artificial Analysis Intelligence Index

AA-Omniscience

AA-Omniscience Index

AA-Omniscience Index

开放性指数

Artificial Analysis 开放性指数：得分

Intelligence Index 比较

Intelligence Index 与每项任务的成本

Cost per Intelligence Index Task

Artificial Analysis Intelligence Index

成本

每项 Intelligence Index 任务的成本

Cost per Intelligence Index Task

运行 Artificial Analysis Intelligence Index 的成本

Cost to Run Artificial Analysis Intelligence Index

价格：缓存命中、输入和输出

Cache Hit

上下文窗口

上下文窗口

Context Window for RAG

Context Window

速度

输出速度

Output Speed

Model Performance Representation

智能指数每项任务耗时

Time per Intelligence Index Task

延迟

延迟: 首个答案 Token 延迟

Time to First Answer Token

端到端响应时间

端到端响应时间

End-to-End Response Time

Model Performance Representation

模型规模（仅开放权重模型）

模型规模：总参数量与活跃参数量

Total Parameters

Active Parameters at Inference Time

常见问题

GLM-4.7 (Reasoning) 何时发布？

谁开发了 GLM-4.7 (Reasoning)？

GLM-4.7 (Reasoning) 有多智能？

GLM-4.7 (Reasoning) 有多快？

GLM-4.7 (Reasoning) 的延迟是多少？

GLM-4.7 (Reasoning) 的费用是多少？

GLM-4.7 (Reasoning) API 的价格是多少？

GLM-4.7 (Reasoning) 是推理模型吗？

GLM-4.7 (Reasoning) 支持哪些输入模态？

GLM-4.7 (Reasoning) 支持哪些输出模态？

GLM-4.7 (Reasoning) 能处理图像吗？

GLM-4.7 (Reasoning) 是多模态模型吗？

GLM-4.7 (Reasoning) 的上下文窗口有多大？

GLM-4.7 (Reasoning) 是开放权重模型吗？

GLM-4.7 (Reasoning) 有多少个参数？

GLM-4.7 (Reasoning) 有多少活跃参数？

GLM-4.7 (Reasoning) 使用什么许可证？

GLM-4.7 (Reasoning) 在基准测试中的表现如何？

可以通过 API 使用 GLM-4.7 (Reasoning) 吗？

可以在哪里使用 GLM-4.7 (Reasoning)？