Supernova:模型智能、性能与价格

Supernova
Supernova

本分析旨在帮助你根据使用场景,选择 Supernova 提供的最佳模型。

最智能

Updated
#1
GLM-5.3 (max) (FP4)GLM-5.3 (max) (FP4)
45
#2
GLM-5.3 (max) (FP4, FAST)GLM-5.3 (max) (FP4, FAST)
45
#3
GLM-5.3 (low) (FP4)GLM-5.3 (low) (FP4)
34
#4
GLM-5.3 (low) (FP4, FAST)GLM-5.3 (low) (FP4, FAST)
34

Intelligence Index

共 4 个模型

速度最快

#1
GLM-5.3 (max) (FP4, FAST)GLM-5.3 (max) (FP4, FAST)
525 t/s
#2
GLM-5.3 (low) (FP4, FAST)GLM-5.3 (low) (FP4, FAST)
473 t/s
#3
GLM-5.3 (max) (FP4)GLM-5.3 (max) (FP4)
224 t/s
#4
GLM-5.3 (low) (FP4)GLM-5.3 (low) (FP4)
208 t/s

输出速度

共 4 个模型

价格最低

#1
GLM-5.3 (max) (FP4)GLM-5.3 (max) (FP4)
$0.90
#2
GLM-5.3 (low) (FP4)GLM-5.3 (low) (FP4)
$0.90
#3
GLM-5.3 (max) (FP4, FAST)GLM-5.3 (max) (FP4, FAST)
$1.80
#4
GLM-5.3 (low) (FP4, FAST)GLM-5.3 (low) (FP4, FAST)
$1.80

每 100 万 token 的混合价格

共 4 个模型

Supernova Labs 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,Supernova Labs 上表现最好的模型是 GLM-5.3 (max) (FP4)(45)、GLM-5.3 (max) (FP4, FAST)(45)和GLM-5.3 (low) (FP4)(34)。
  • 输出速度方面,最快的模型是 GLM-5.3 (max) (FP4, FAST)(525 t/s)、GLM-5.3 (low) (FP4, FAST)(473 t/s)和GLM-5.3 (max) (FP4)(224 t/s)。 各模型之间的速度差异显著,最快与最慢相差 153%。
  • 延迟方面,GLM-5.3 (max) (FP4, FAST)(5.29 秒)、GLM-5.3 (low) (FP4, FAST)(5.59 秒)和GLM-5.3 (max) (FP4)(10.62 秒) 的首个答案 Token 延迟最低。
  • 价格方面,GLM-5.3 (max) (FP4)($0.90)、GLM-5.3 (low) (FP4)($0.90)和GLM-5.3 (max) (FP4, FAST)($1.80) 每 100 万 token 的混合价格最低。
  • 上下文窗口方面,GLM-5.3 (max) (FP4)(1M)、GLM-5.3 (max) (FP4, FAST)(1M)和GLM-5.3 (low) (FP4)(1M) 支持 Supernova Labs 上最大的上下文窗口。
  • GLM-5.3 (max) (FP4) 在智能与成本效益之间取得了最佳平衡。若追求最快输出,GLM-5.3 (max) (FP4, FAST) 是首选。

亮点

Updated
Artificial Analysis Intelligence Index · Higher is better
Output tokens per second · Higher is better
USD per 1M tokens (blended) · Lower is better

智能评测

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

Intelligence Evaluations

Intelligence evaluations measured independently by Artificial Analysis · Higher is better
See more

Agentic knowledge work, (Elo-500)/2000

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

Agentic SaaS workflows

Agentic coding & terminal use

Coding

Reasoning & knowledge

Professional document reasoning, All-pass

CritPtUnder review

Physics reasoning

Long context reasoning

Legal agentic work, criterion pass rate

No data available

Agentic business operations

Agentic scientific research workflows in a terminal

Quantitative analysis on spreadsheets & documents

No data available

Agentic tool use

Kubernetes incident root-cause analysis

Visual reasoning

No data available

Medical long context reasoning

Intelligence Index vs. Price

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

上下文窗口

Context Window

Context window: tokens limit · Higher is better

价格

Intelligence Index vs. Price

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

性能摘要

Output Speed vs. Price

Output speed: output tokens per second · USD per 1M tokens (blended)
Most attractive quadrant
Pareto line

速度

按输出速度(每秒 token 数)衡量

Output Speed

Output tokens per second · Higher is better

延迟

按首 Token 延迟(秒)衡量

Latency: Time To First Answer Token

Seconds to first answer token received · Accounts for reasoning model 'thinking' time

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

End-to-End Response Time vs. Price

End-to-end response time: end-to-end seconds to output 500 tokens · USD per 1M tokens (blended)
Most attractive quadrant
Pareto line

进一步分析
Z AI 标志
GLM-5.3 (max) (FP4)
1M
开放
45
$2.64
224
1.70
12.85
8.92
Z AI 标志
GLM-5.3 (max) (FP4, FAST)
1M
开放
45
$5.29
525
1.49
6.25
3.81
Z AI 标志
GLM-5.3 (low) (FP4)
1M
开放
34
$1.12
208
1.78
13.83
9.63
Z AI 标志
GLM-5.3 (low) (FP4, FAST)
1M
开放
34
$2.24
473
1.36
6.65
4.23

关键定义

常见问题

关于 Supernova 的常见问题