o1
OpenAIOpenAI o-seriesProprietary
描述
A research preview model focused on mathematical and logical reasoning capabilities, demonstrating improved performance on tasks requiring step-by-step reasoning, mathematical problem-solving, and code generation. The model shows enhanced capabilities in formal reasoning while maintaining strong general capabilities.
发布日期
2024-12-05
参数规模
—
上下文长度
200K
支持模态
image, pdf, text
能力雷达图
39
general
49
coding
80
reasoning
48
science
60
agents
70
multimodal
排行榜排名
基准测试分数 (LLM Stats)
(LLM Stats (zeroeval))Biology
GPQANYU + Cohere + Anthropic (2023)
78.0%自报
GPQA Biology
69.2%自报
Chemistry
GPQA Chemistry
64.7%自报
Code
HumanEvalOpenAI (2021)
88.1%自报
SWE-Bench Verified
41.0%自报
Communication
TAU-bench Retail
70.8%自报
TAU-bench Airline
50.0%自报
Factuality
SimpleQA
47.0%自报
Finance
MMLU
91.8%自报
General
MMMLU
87.7%自报
MMMU
77.6%自报
LiveBench
67.0%自报
Math
GSM8k
97.1%自报
MATH
96.4%自报
MGSM
89.3%自报
AIME 2024
74.3%自报
MathVista
71.8%自报
FrontierMath
5.5%自报
Physics
GPQA Physics
92.8%自报
AA 评测指数
(Artificial Analysis)Coding Index(Artificial Analysis)39.7
Intelligence Index(Artificial Analysis)23.9
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
Gpqa(NYU + Cohere + Anthropic (2023))0.7
Aime(MAA (Mathematical Association of America))0.7
Ifbench(Google Research (2023))0.7
Livecodebench(UC Berkeley + MIT + Cornell (2024))0.7
Lcr(Artificial Analysis)0.6
Tau2(Sierra + U Toronto + Vector Institute (2025))0.6
Scicode(UIUC + Argonne National Lab (2024))0.4
Terminalbench Hard(Stanford × Laude Institute (2026))0.1
Hle(Center for AI Safety + Scale AI (2025))0.1
LLM Stats 分类评分
(LLM Stats (zeroeval))Language90
Legal90
Finance90
Math80
Physics80
Healthcare80
Biology80
Chemistry80
Multimodal70
Reasoning70
General70
Vision70
Code60
Communication60
Tool Calling60
Factuality50
Frontend Development40
定价
输入价格$15 / 1M tokens
输出价格$60 / 1M tokens
混合价格(3:1)$26.25 / 1M tokens
缓存读取价格$7.5 / 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
供应商价格排行
13 个供应商
最便宜: Poe最贵: Impossibl
供应商输入输出
1Poe最便宜
$14
$54
2OpenAI主要
$15
$60
3NanoGPT
$15
$60
4OpenRouter
$15
$60
5Kilo Gateway
$15
$60
6Cloudflare AI Gateway
$15
$60
7Helicone
$15
$60
8Azure Cognitive Services
$15
$60
9Vercel AI Gateway
$15
$60
10LLM Gateway
$15
$60
11Azure
$15
$60
12Merge Gateway
$15
$60
13Impossibl
$15
$60
比较该模型在不同 API 供应商之间的定价。