See key specs and per-benchmark scores for each model/mode. Scroll horizontally for all columns. 当前对比 2 个模型的评测数据与核心参数。
Each axis is a category average, normalized to a 100-point radar.
Relative edge: 综合评估 +23.4 / Relative gap: 多模态理解 -3.8
Relative edge: 多模态理解 +3.8 / Relative gap: 综合评估 -23.4
Method: for each model and benchmark, the chart first averages all scores in the current mode scope instead of taking the best score, then averages those benchmark scores within each category. Only benchmarks with at least two selected models scored are included; missing values are not counted as zero.
Best overall
GPT-5.2 · 63.01
Best single
GPT-5.2 · τ²-Bench - Telecom 98.70
Modality coverage
GPT-5.2 · 2 modalities
Head to head
9
Benchmarks
8
Wins
1
Losses
+7.58
Average diff
Compare benchmark results across thinking modes and tool usage.
Data sourced primarily from official releases (GitHub, Hugging Face, papers), then benchmark leaderboards, then third-party evaluators. Learn about our data methodology
Complete scores for each model/mode across selected benchmarks.
9 benchmarks with comparable scores. Each model shows its best score; mode label is displayed below.
| Benchmark | GPT-5.2 | GPT-5 |
|---|---|---|
HLE 综合评估 | 45.50Deep Thinking Mode ｜ Tools | 35.20Thinking Enabled ｜ Tools |
SWE-Bench Pro - Public 编程与软件工程 | 55.60Thinking Level · Extra High ｜ Tools | 36.30Thinking Level · High |
SWE-bench Verified 编程与软件工程 | 80.00Thinking Level · Extra High ｜ Tools | 72.80Thinking Level · High |
FrontierMath 数学推理 | 40.30Thinking Level · Extra High ｜ Tools | 26.30Thinking Level · High ｜ Tools |
18.80Thinking Level · Extra High | 12.50Thinking Level · High | |
MMMU 多模态理解 | 80.40Thinking Level · Extra High ｜ Tools | 84.20Thinking Level · High |
τ²-Bench Agent能力评测 | 82.00Thinking Level · Extra High ｜ Tools | 80.00Thinking Enabled ｜ Tools |
τ²-Bench - Telecom Agent能力评测 | 98.70Thinking Level · Extra High ｜ Tools | 96.70Thinking Level · High ｜ Tools |
BrowseComp AI Agent - 信息收集 | 65.80Thinking Level · Extra High ｜ Tools | 54.90Thinking Enabled ｜ Tools |
Side-by-side input/output token pricing
Licensing, MoE architecture, and multi-modality support.
| Features & specs | GPT-5.2OpenAI | GPT-5OpenAI |
|---|---|---|
Core specsRelease | 2025-12-11 | 2025-08-07 |
Context length | 400K | 400K |
Max output | Not provided | 131072 |
MoE | No | No |
Supported modes | No mode data | 常规模式（Non-Thinking Mode）思考模式（Thinking Mode）深度思考（Deeper Thinking Mode） |
LicenseCode Open Source | Not provided | Not provided |
Weights Open Source | Not provided | Not provided |
Commercial use | 不开源 | 不开源 |
Modality supportText Input/Output | / | / |
Image Input/Output | / | / |
ResourcesPaper / report | Introducing GPT-5.2 | Introducing GPT-5 |
DataLearner blog | Not provided | OpenAI发布GPT-5：这是一个包含实时路由的AI系统，而不仅仅是一个模型 |
GPT-5
OpenAI
