GPT-5 (High)
OpenAIGPTProprietary
Описание
GPT-5 is a flagship model from OpenAI designed for coding, reasoning, and agentic tasks across domains. It is optimized for coding and agentic tasks with higher reasoning capabilities and medium speed.
Дата выхода
2025-08-07
Параметры
—
Длина контекста
400K
Модальности
image, text
Радар способностей
43
general
56
coding
95
reasoning
68
science
80
agents
90
multimodal
Рейтинги
| Домен | #Место | Оценка | Источник |
|---|---|---|---|
| Рейтинг кодинга | 215 | 64.0 | AA |
| Общий рейтинг | 87 | 69.0 | AA |
| Мультимодальный рейтинг | 20 | 63.0 | LS |
| Наука | 152 | 66.0 | AA |
Оценки бенчмарков (LLM Stats)
(LLM Stats (zeroeval))Chat
Multi-Challenge
69.6%Сам.
Tau2 Airline
62.6%Сам.
Communication
Tau2 Telecom
96.7%Сам.
Tau2 Retail
81.1%Сам.
General
MMLU
92.5%Сам.
Aider-Polyglot
88.0%Сам.
Internal API instruction following (hard)
64.0%Сам.
LongFact Objects
0.8%Сам.
LongFact Concepts
0.7%Сам.
Healthcare
HealthBench Hard
1.6%Сам.
Language
COLLIE
99.0%Сам.
Long Context
OpenAI-MRCR: 2 needle 128k
95.2%Сам.
OpenAI-MRCR: 2 needle 256k
86.8%Сам.
Math
AIME 2025
94.6%Сам.
HMMT 2025
93.3%Сам.
MATH
84.7%Сам.
FrontierMath
26.3%Сам.
Multimodal
VideoMMMU
84.6%Сам.
MMMU
84.2%Сам.
Reasoning
SWE-Lancer (IC-Diamond subset)
100.0%Сам.
HumanEvalOpenAI (2021)
93.4%Сам.
BrowseComp Long Context 128k
90.0%Сам.
BrowseComp Long Context 256k
88.8%Сам.
GPQANYU + Cohere + Anthropic (2023)
87.3%Сам.
CharXiv-R
81.1%Сам.
Graphwalks BFS <128k
78.3%Сам.
SWE-Bench Verified
74.9%Сам.
Graphwalks parents <128k
73.3%Сам.
BrowseCompOpenAI (2025)
54.9%Сам.
Humanity's Last Exam
24.8%Сам.
FActScoreMin et al. (NYU/UW, 2023)
1.0%Сам.
Vision
VideoMME w sub.
86.7%Сам.
MMMU-Pro
78.4%Сам.
ERQA
65.7%Сам.
Индексы оценки AA
(Artificial Analysis)Math 500(OpenAI (2024), subset of Hendrycks et al. MATH (2021))99.4
Aime(MAA (Mathematical Association of America))95.7
Aime 25(MAA (Mathematical Association of America))94.3
Math Index(Artificial Analysis)94.3
Mmlu Pro(TIGER-Lab (Univ. of Waterloo, Toronto, CMU, 2024))87.1
Gpqa(NYU + Cohere + Anthropic (2023))85.4
Tau2(Sierra + U Toronto + Vector Institute (2025))84.8
Livecodebench(UC Berkeley + MIT + Cornell (2024))84.6
Lcr(Artificial Analysis)78.2
Ifbench(Google Research (2023))73.1
Coding Index(Artificial Analysis)37.8
Terminalbench V2 135.2
Terminalbench Hard(Stanford × Laude Institute (2026))32.6
Hle(Center for AI Safety + Scale AI (2025))28.5
Intelligence Index(Artificial Analysis)23.0
Tau Banking22.1
Оценки категорий LLM Stats
(LLM Stats (zeroeval))Spatial Reasoning7
Vision4
Reasoning2
General1
Language100
Long Context100
Writing100
Legal90
Physics90
Finance90
Biology90
Chemistry90
Code90
Video90
Multimodal80
Communication80
Tool Calling80
Chat70
Math70
Search70
Frontend Development70
Healthcare70
Structured Output60
Agents50
Цены
Цена ввода$1.25 / 1M токенов
Цена вывода$10 / 1M токенов
Смешанная цена (3:1)$3.438 / 1M токенов
Цена чтения кэша$0.125 / 1M токенов
Скорость
Токенов/сек0.0
Задержка первого токена0.00s
Время до первого ответа0.00s
Рейтинг цен провайдеров
Рейтинг цен провайдеров
1 провайдеров
ПровайдерВводВывод
1OpenAIОсновной
$1.25
$10
Сравнение цен разных API-провайдеров для этой модели.