跳转到主要内容

GPT-5.5 (medium)

OpenAIGPT

描述

GPT-5.5 is OpenAI's smartest model yet, designed for real work across agentic coding, computer use, knowledge work, and early scientific research. It matches GPT-5.4 per-token latency in real-world serving while reaching a much higher level of intelligence and using significantly fewer tokens to complete the same tasks. GPT-5.5 supports a 1M-token context window in the API and a 400K-token context window in Codex, with state-of-the-art results on Terminal-Bench 2.0, OSWorld-Verified, GDPval, FrontierMath, and CyberGym.

发布日期
2026-04-23
参数规模
上下文长度
1.1M
支持模态
image, pdf, text

能力雷达图

48
general
69
coding
93
reasoning
69
science估算
80
agents
85
multimodal

Science 在缺少专门科学评测时使用推理能力代理估算。

排行榜排名

领域#排名分数来源
代码能力榜4
94.0
AA
通用能力榜8
86.0
AA
科学能力8
88.0
AA

基准测试分数 (LLM Stats)

Agents

GDPval-AA1135.00 / 3000自报
BrowseComp84.4%自报
Terminal-Bench 2.082.7%自报
CyberGym81.8%自报
BixBench80.5%自报
OSWorld-Verified78.7%自报
MCP Atlas75.3%自报
FrontierSWE73.0%自报
Finance Agent60.0%自报
SWE-Bench Pro58.6%自报
Toolathlon55.6%自报
OfficeQA Pro54.1%自报
Finance Agent v251.8%自报
GeneBench25.0%自报
Legal Agent Benchmark2.1%自报

Biology

GPQA93.6%自报

Communication

Tau2 Telecom98.0%自报

Finance

GDPval-MM84.9%自报

General

MMMU-Pro83.2%自报
LiveBench80.7%自报
MRCR v2 (8-needle)74.0%自报

Long Context

Graphwalks parents >128k58.5%自报
Graphwalks BFS >128k45.4%自报

Math

Humanity's Last Exam52.2%自报
FrontierMath35.4%自报

Reasoning

ARC-AGI95.0%自报
ARC-AGI v285.0%自报

AA 评测指数

Coding Index
71.5
Intelligence Index
50.4
Gpqa
0.9
Tau2
0.9
Terminalbench V2 1
0.8
Lcr
0.7
Ifbench
0.7
Terminalbench Hard
0.6
Scicode
0.5
Hle
0.4
Tau Banking
0.3

LLM Stats 分类评分

Legal
100
Finance
100
General
100
Agents
88
Reasoning
52
Communication
100
Physics
90
Biology
90
Chemistry
90
Multimodal
80
Safety
80
Search
80
Tool Calling
80
Vision
80
Spatial Reasoning
70
Code
70
Long Context
60
Math
60

定价

输入价格$5 / 1M tokens
输出价格$30 / 1M tokens
混合价格(3:1)$11.25 / 1M tokens
缓存读取价格$0.5 / 1M tokens

速度

Tokens/秒70.3
首Token延迟6.92s
首回答延迟6.92s

供应商价格排行

供应商价格排行

1 个供应商

供应商输入输出
1OpenAI主要
$5
$30

比较该模型在不同 API 供应商之间的定价。

外部链接