GPT-5.2 (Xhigh)
OpenAIGPTProprietary
描述
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.
發布日期
2025-12-11
參數規模
—
上下文長度
400K
支援模態
image, text
能力雷達圖
80
general
70
coding
70
reasoning
77
science估算
70
agents
85
multimodal
缺少專門科學評測時,Science 由 LLM Stats 科學得分或推理能力估算。
排行榜排名
基準測試分數 (LLM Stats)
(LLM Stats (zeroeval))Agents
Toolathlon
46.3%自報
Communication
Tau2 Telecom
98.7%自報
Tau2 Retail
82.0%自報
Language
MMMLU
89.6%自報
Math
AIME 2025
100.0%自報
HMMT 2025
99.4%自報
LiveBench
74.8%
FrontierMath
40.3%自報
Multimodal
VideoMMMU
85.9%自報
Reasoning
Graphwalks BFS <128k
94.0%自報
GPQANYU + Cohere + Anthropic (2023)
92.4%自報
BrowseComp Long Context 128k
92.0%自報
BrowseComp Long Context 256k
89.8%自報
Graphwalks parents <128k
89.0%自報
ARC-AGI
86.2%自報
CharXiv-R
82.1%自報
SWE-Bench Verified
80.0%自報
SWE-Lancer (IC-Diamond subset)
74.6%自報
BrowseCompOpenAI (2025)
65.8%自報
MCP Atlas
60.6%自報
ARC-AGI v2
52.9%自報
Humanity's Last Exam
34.5%自報
Vision
ScreenSpot Pro
86.3%自報
MMMU-Pro
79.5%自報
AA 評測指數
(Artificial Analysis)暫無 AA 評測資料
LLM Stats 分類評分
(LLM Stats (zeroeval))Language90
Physics90
Grounding90
Healthcare90
Biology90
Chemistry90
Communication90
Multimodal80
Reasoning80
Search80
Spatial Reasoning80
Frontend Development80
General80
Math70
Code70
Tool Calling70
Vision70
Agents60
定價
輸入價格$1.75 / 1M tokens
輸出價格$14 / 1M tokens
混合價格(3:1)$4.8125 / 1M tokens
快取讀取價格$0.175 / 1M tokens
速度
Tokens/秒72.1
首Token延遲99.14s
首回答延遲99.14s
供應商價格排行
供應商價格排行
2 個供應商
最便宜: OpenAI最貴: Neon
供應商輸入輸出
1OpenAI主要
$1.75
$14
2Neon
$1.75
$14
比較該模型在不同 API 供應商之間的定價。