o3-mini

OpenAIOpenAI o-seriesProprietary

描述

A smaller variant of O3, expected to offer enhanced multimodal capabilities, improved reasoning, and more efficient resource utilization compared to previous models while maintaining strong performance on core tasks.

发布日期

2025-01-31

参数规模

—

上下文长度

200K

支持模态

text

能力雷达图

general

coding

reasoning

science

agents

multimodal

排行榜排名

领域	#排名	分数	来源
代码能力榜	259	44.0	AA
通用能力榜	284	44.0	AA
科学能力	217	52.0	AA

基准测试分数 (LLM Stats)

(LLM Stats (zeroeval))

Biology

GPQANYU + Cohere + Anthropic (2023)

77.2%自报

Code

Aider-Polyglot

66.7%自报

Aider-Polyglot Edit

60.4%自报

SWE-Bench Verified

49.3%自报

SWE-Lancer

18.0%自报

SWE-Lancer (IC-Diamond subset)

7.4%自报

Communication

Multi-IF

79.5%自报

TAU-bench Retail

57.6%自报

Multi-Challenge

39.9%自报

TAU-bench Airline

32.4%自报

Factuality

SimpleQA

15.0%自报

Finance

MMLU

86.9%自报

General

IFEvalGoogle Research (2023)

93.9%自报

LiveBench

84.6%自报

Multilingual MMLU

80.7%自报

Internal API instruction following (hard)

50.0%自报

Language

COLLIE

98.7%自报

Long Context

OpenAI-MRCR: 2 needle 128k

18.7%自报

ComplexFuncBench

17.6%自报

Math

MATH

97.9%自报

MGSM

92.0%自报

AIME 2024

87.3%自报

FrontierMath

9.2%自报

Reasoning

Graphwalks parents <128k

58.3%自报

Graphwalks BFS <128k

51.0%自报

AA 评测指数

(Artificial Analysis)

Intelligence Index(Artificial Analysis)

19.2

Math 500(OpenAI (2024), subset of Hendrycks et al. MATH (2021))

1.0

Mmlu Pro(TIGER-Lab (Univ. of Waterloo, Toronto, CMU, 2024))

0.8

Aime(MAA (Mathematical Association of America))

0.8

Gpqa(NYU + Cohere + Anthropic (2023))

0.7

Livecodebench(UC Berkeley + MIT + Cornell (2024))

0.7

Scicode(UIUC + Argonne National Lab (2024))

0.4

Tau2(Sierra + U Toronto + Vector Institute (2025))

0.3

Hle(Center for AI Safety + Scale AI (2025))

0.1

Terminalbench Hard(Stanford × Laude Institute (2026))

0.1

LLM Stats 分类评分

(LLM Stats (zeroeval))

Writing

100

Legal

Instruction Following

Language

Finance

Healthcare

Math

Physics

Biology

Chemistry

Reasoning

Structured Output

General

Spatial Reasoning

Frontend Development

Communication

Code

Tool Calling

Long Context

Factuality

定价

输入价格$1.1 / 1M tokens

输出价格$4.4 / 1M tokens

混合价格(3:1)$1.925 / 1M tokens

缓存读取价格$0.55 / 1M tokens

速度

Tokens/秒0.0

首Token延迟0.00s

首回答延迟0.00s

供应商价格排行

1 个供应商

供应商输入输出

1OpenAI主要

$1.1

$4.4

比较该模型在不同 API 供应商之间的定价。

外部链接

LLM Stats Artificial Analysis