GPT-5.2 (xhigh)
OpenAIGPTProprietary
Описание
GPT‑5.2 introduces substantial gains in professional knowledge work, outperforming experts on GDPval with 70.9% wins or ties, and setting new highs in coding (SWE‑Bench Pro 55.6%), science (GPQA Diamond ~92–93%), math (AIME 2025: 100%), long‑context accuracy up to 256k tokens, and reliable tool‑calling (Tau2 Telecom 98.7%). It rolls out as Instant, Thinking, and Pro—faster, more structured, and less error‑prone—priced at $1.75/1M input and $14/1M output tokens, with Pro variants supporting xhigh reasoning for top‑quality, end‑to‑end execution.
Дата выхода
2025-12-11
Параметры
—
Длина контекста
400K
Модальности
image, text
Радар способностей
56
general
81
coding
98
reasoning
66
science
70
agents
85
multimodal
Рейтинги
| Домен | #Место | Оценка | Источник |
|---|---|---|---|
| Агентные возможности | 101 | 39.0 | LS |
| Рейтинг кодинга | 36 | 88.0 | AA |
| Общий рейтинг | 36 | 82.0 | AA |
| Мультимодальный рейтинг | 32 | 55.0 | LS |
| Наука | 42 | 83.0 | AA |
Оценки бенчмарков (LLM Stats)
(LLM Stats (zeroeval))Agents
BrowseCompOpenAI (2025)
65.8%Сам.
MCP Atlas
60.6%Сам.
Toolathlon
46.3%Сам.
Biology
GPQANYU + Cohere + Anthropic (2023)
92.4%Сам.
Code
SWE-Bench Verified
80.0%Сам.
SWE-Lancer (IC-Diamond subset)
74.6%Сам.
Communication
Tau2 Telecom
98.7%Сам.
Tau2 Retail
82.0%Сам.
General
MMMLU
89.6%Сам.
MMMU-Pro
79.5%Сам.
LiveBench
74.8%
Grounding
ScreenSpot Pro
86.3%Сам.
Healthcare
VideoMMMU
85.9%Сам.
Math
AIME 2025
100.0%Сам.
HMMT 2025
99.4%Сам.
FrontierMath
40.3%Сам.
Humanity's Last Exam
34.5%Сам.
Multimodal
CharXiv-R
82.1%Сам.
Reasoning
Graphwalks BFS <128k
94.0%Сам.
BrowseComp Long Context 128k
92.0%Сам.
BrowseComp Long Context 256k
89.8%Сам.
Graphwalks parents <128k
89.0%Сам.
ARC-AGI
86.2%Сам.
ARC-AGI v2
52.9%Сам.
Индексы оценки AA
(Artificial Analysis)Math Index(Artificial Analysis)99.0
Intelligence Index(Artificial Analysis)43.3
Aime 25(MAA (Mathematical Association of America))1.0
Gpqa(NYU + Cohere + Anthropic (2023))0.9
Livecodebench(UC Berkeley + MIT + Cornell (2024))0.9
Mmlu Pro(TIGER-Lab (Univ. of Waterloo, Toronto, CMU, 2024))0.9
Tau2(Sierra + U Toronto + Vector Institute (2025))0.8
Lcr(Artificial Analysis)0.8
Ifbench(Google Research (2023))0.8
Scicode(UIUC + Argonne National Lab (2024))0.5
Terminalbench Hard(Stanford × Laude Institute (2026))0.5
Hle(Center for AI Safety + Scale AI (2025))0.4
Оценки категорий LLM Stats
(LLM Stats (zeroeval))Physics90
Language90
Grounding90
Healthcare90
Biology90
Chemistry90
Communication90
Multimodal80
Reasoning80
Search80
Spatial Reasoning80
Frontend Development80
General80
Math70
Code70
Tool Calling70
Vision70
Agents60
Цены
Цена ввода$1.75 / 1M токенов
Цена вывода$14 / 1M токенов
Смешанная цена (3:1)$4.813 / 1M токенов
Цена чтения кэша$0.175 / 1M токенов
Скорость
Токенов/сек0.0
Задержка первого токена0.00s
Время до первого ответа0.00s
Рейтинг цен провайдеров
Рейтинг цен провайдеров
2 провайдеров
Самый дешевый: OpenAIСамый дорогой: Neon
ПровайдерВводВывод
1OpenAIСамый дешевый
$0
$0.00001
2Neon
$1.75
$14
Сравнение цен разных API-провайдеров для этой модели.