Multiverse Computing:模型智能、性能与价格

Multiverse Computing
Multiverse Computing

本分析旨在帮助你根据使用场景,选择 Multiverse Computing 提供的最佳模型。

最智能

#1
GLM-5.3 (max)GLM-5.3 (max)
32
#2
Qwen3.8 27B (xhigh)Qwen3.8 27B (xhigh)
20
#3
Qwen3.8 27B (non-reasoning)Qwen3.8 27B (non-reasoning)
10

Intelligence Index

共 4 个模型

每任务耗时最短

#1
Qwen3.8 27B (non-reasoning)Qwen3.8 27B (non-reasoning)
3.4 min
#2
Qwen3.8 27B (xhigh)Qwen3.8 27B (xhigh)
7.2 min
#3
GLM-5.3 (max)GLM-5.3 (max)
18.5 min

每个 Intelligence Index 任务耗时(分钟)

共 4 个模型

价格最低

#1
Qwen3.8 27B (xhigh)Qwen3.8 27B (xhigh)
$0.07
#2
Qwen3.8 27B (non-reasoning)Qwen3.8 27B (non-reasoning)
$0.07
#3
GLM-5.2 (max)GLM-5.2 (max)
$0.10
#4
GLM-5.3 (max)GLM-5.3 (max)
$0.12

每 100 万 token 的混合价格

共 4 个模型

Multiverse Computing 提供 4 个模型,每个模型的智能、性能和价格特征各不相同。 下方对比了各模型的关键指标。

  • 智能方面,Multiverse Computing 上表现最好的模型是 GLM-5.3 (max)(32)、Qwen3.8 27B (xhigh)(20)和Qwen3.8 27B (non-reasoning)(10)。
  • 每任务耗时方面,Qwen3.8 27B (non-reasoning)(3.4 分钟)、Qwen3.8 27B (xhigh)(7.2 分钟)和GLM-5.3 (max)(18.5 分钟) 在 Multiverse Computing 上完成一个 Intelligence Index 任务最快。
  • 延迟方面,Qwen3.8 27B (non-reasoning)(1.65 秒)、GLM-5.2 (max)(19.27 秒)和Qwen3.8 27B (xhigh)(19.71 秒) 的首个答案 Token 延迟最低。
  • 价格方面,Qwen3.8 27B (xhigh)($0.07)、Qwen3.8 27B (non-reasoning)($0.07)和GLM-5.2 (max)($0.10) 每 100 万 token 的混合价格最低。
  • 上下文窗口方面,GLM-5.3 (max)(1M)、GLM-5.2 (max)(1M)和Qwen3.8 27B (xhigh)(262k) 支持 Multiverse Computing 上最大的上下文窗口。
Artificial Analysis Intelligence Index · Higher is better
Inference time (minutes) per Intelligence Index task · Lower is better
USD per 1M tokens (blended) · Lower is better

智能

Artificial Analysis Intelligence Index

Artificial Analysis Intelligence Index v5.0.1 incorporates 9 evaluations: AA-Briefcase v1.1, AutomationBench-AA, Terminal-Bench 4.0, FrontierCode-AA 2.0, AA-Omniscience v1.1, GDP.pdf, HLE-Diamond, CritPt v2 (Hard Subset), and Terminal-Bench-Science 0.1

Evaluations

Evaluations measured independently by Artificial Analysis · AA-PersonalityBench scores are percentile ranks from 1 to 100
查看更多

Agentic knowledge work, Elo

Agentic SaaS workflows

Agentic scientific research workflows in a terminal

Agentic software engineering in open-source repositories

Reasoning & knowledge

Professional document reasoning, All-pass

Agentic coding & terminal use

Physics reasoning

Quantitative analysis on spreadsheets & documents

Legal agentic work, Hallucination-Gated All-Pass Rate

Agentic business operations

Agentic real-world work tasks, Elo

Agentic interactive 3D experiences, Elo

Product and brand videos designed in code, Elo

Kubernetes incident root-cause analysis

Medical long context reasoning

Visual reasoning

Rich, varied, well-crafted language

Well-structured, scannable, answer-first

Truthful, acknowledges uncertainty, non-sycophantic

Use of filler and generic language

Intelligence Index 与价格

Blended at 7:2:1 (cache-input-output) · USD per 1M tokens (blended)
Most attractive quadrant
Pareto line

价格

每项 Intelligence Index 任务的成本

Weighted average cost (USD) per Intelligence Index task · Segmented by token type · Lower is better

Intelligence Index 与价格

Blended at 7:2:1 (cache-input-output) · USD per 1M tokens (blended)
Most attractive quadrant
Pareto line

速度

已测量的单次请求和多轮智能体工作负载的平均值

Intelligence Index 每项任务耗时

Output tokens divided by end-to-end output speed (minutes per Intelligence Index task) · Lower is better

质量

Endpoint Accuracy Index

各接入点准确率相对其模型基准接入点的百分比 · 颜色表示差距是否具有统计显著性

延迟

已测量的单次请求和多轮智能体工作负载的平均值

首个答案 Token 延迟

经过推理后首个答案 token 到达所需的秒数 · 已测量的单次请求和多轮智能体工作负载的平均值 · 越低越好

可靠性

在所有已测量的单次请求和多轮智能体工作负载中,接入点失败的请求占比

接入点错误率

接入点因错误、限速或超时而失败的请求占比 · 涵盖所有已测量的单次请求和多轮智能体工作负载 · 越低越好

缓存

缓存命中率

多轮智能体会话各次运行在最近四周的中位数,更新于 2026年10月26日 · 越高越好

每任务成本与缓存命中率

Weighted average cost (USD) per Intelligence Index task · Cache hit rate: median of the last four weeks, updated Oct 26, 2026
Most attractive quadrant
Pareto line

进一步分析
Z AI 标志
GLM-5.3 (max)
32
$1.48
20.5
55.89
0.0%
1M
开放
Alibaba 标志
Qwen3.8 27B (xhigh)
20
$1.47
8.3
25.72
0.7%
262k
开放
Alibaba 标志
Qwen3.8 27B (non-reasoning)
10
$1.95
3.4
1.63
0.0%
262k
开放
Mistral 标志
Mistral Small 3.1
--
--
--
--
--
128k
开放
Z AI 标志
GLM-5.2 (max)
--
--
--
13.09
0.0%
1M
开放

关键定义

常见问题

关于 Multiverse Computing 的常见问题