GPT-5.2vsGemini 3.0 Pro (Preview 11-2025)
Across 15 shared benchmarks, GPT-5.2 leads overall: GPT-5.2 wins 8, Gemini 3.0 Pro (Preview 11-2025) wins 6, with 1 ties and an average score difference of +0.29.
GPT-5.2
OpenAI · 2025-12-11 · Chat model
Gemini 3.0 Pro (Preview 11-2025)
Google Deep Mind · 2025-11-18 · Multimodal model
GPT-5.28 wins(53%)Ties1(40%)6 winsGemini 3.0 Pro (Preview 11-2025)
Benchmark scores
Grouped by capability, sorted by largest gap within each. 15 shared benchmarks.
General Knowledge
Gemini 3.0 Pro (Preview 11-2025) 3/5| Benchmark | GPT-5.2 | Gemini 3.0 Pro (Preview 11-2025) | Diff |
|---|---|---|---|
| LiveBench | 48.9194 / 115Normal (No Tools) | 73.3924 / 115Thinking High (No Tools) | -24.48 |
| ARC-AGI-2 | 54.2023 / 62深度思考（无工具、并行） | 45.1026 / 62 | +9.10 |
| ARC-AGI | 90.5017 / 68深度思考（无工具、并行） | 87.5020 / 68 | +3 |
| GPQA Diamond | 93.209 / 187深度思考（无工具、并行） | 93.805 / 187 | -0.60 |
| HLE | 45.5041 / 172Deep Thinking (With Tools + Internet) | 45.8040 / 172 | -0.30 |
Math and Reasoning
GPT-5.2 2/3| Benchmark | GPT-5.2 | Gemini 3.0 Pro (Preview 11-2025) | Diff |
|---|---|---|---|
| AIME2025 | 1001 / 107极高强度思考（无工具） | 9525 / 107 | +5 |
| FrontierMath | 40.308 / 60极高强度思考（工具） | 3810 / 60 | +2.30 |
| FrontierMath - Tier 4 | 18.8016 / 80Thinking High (No Tools) | 18.8016 / 80 | — |
Agent Level Benchmark
Even 2/2| Benchmark | GPT-5.2 | Gemini 3.0 Pro (Preview 11-2025) | Diff |
|---|---|---|---|
| τ²-Bench | 8212 / 43极高强度思考（工具） | 85.408 / 43 | -3.40 |
| τ²-Bench - Telecom | 98.704 / 35极高强度思考（工具） | 985 / 35 | +0.70 |
AI Agent - Information Search
GPT-5.2 1/1| Benchmark | GPT-5.2 | Gemini 3.0 Pro (Preview 11-2025) | Diff |
|---|---|---|---|
| BrowseComp | 65.8031 / 53Deep Thinking (With Tools + Internet) | 59.2038 / 53 | +6.60 |
AI Agent - Tool Usage
Gemini 3.0 Pro (Preview 11-2025) 1/1| Benchmark | GPT-5.2 | Gemini 3.0 Pro (Preview 11-2025) | Diff |
|---|---|---|---|
| MCP-Atlas | 67.6018 / 27极高强度思考（工具） | 70.3015 / 27Normal (With Tools) | -2.70 |
Coding and Software Engineer
GPT-5.2 1/1| Benchmark | GPT-5.2 | Gemini 3.0 Pro (Preview 11-2025) | Diff |
|---|---|---|---|
| SWE-bench Verified | 8017 / 112极高强度思考（工具） | 76.2036 / 112 | +3.80 |
Commonsense Reasoning
Gemini 3.0 Pro (Preview 11-2025) 1/1| Benchmark | GPT-5.2 | Gemini 3.0 Pro (Preview 11-2025) | Diff |
|---|---|---|---|
| Simple Bench | 45.8033 / 63Thinking High (No Tools) | 76.405 / 63Thinking (No Tools) | -30.60 |
Productivity Knowledge
GPT-5.2 1/1| Benchmark | GPT-5.2 | Gemini 3.0 Pro (Preview 11-2025) | Diff |
|---|---|---|---|
| GDPval-AA | 70.909 / 21Thinking High (With Tools) | 3518 / 21 | +35.90 |
Specs
| Field | GPT-5.2 | Gemini 3.0 Pro (Preview 11-2025) |
|---|---|---|
| Publisher | OpenAI | Google Deep Mind |
| Release date | 2025-12-11 | 2025-11-18 |
| Model type | Chat model | Multimodal model |
| Architecture | Dense | Dense |
| Parameters | Not available | Not available |
| Context length | 400K | 1000K |
| Max output | Not available | 64K |
API pricing
Prices use DataLearner records when available; missing fields are not inferred.
| Item | GPT-5.2 | Gemini 3.0 Pro (Preview 11-2025) |
|---|---|---|
| Text input | $1.75 / 1M tokens | $2 / 1M tokens |
| Text output | $14 / 1M tokens | $12 / 1M tokens |
| Cache read | $0.175 / 1M tokens | Not public |
| Cache write | $1.75 / 1M tokens | Not public |
Summary
- GPT-5.2leads in:Math and Reasoning (2/3), AI Agent - Information Search (1/1), Coding and Software Engineer (1/1), Productivity Knowledge (1/1)
- Gemini 3.0 Pro (Preview 11-2025)leads in:General Knowledge (3/5), AI Agent - Tool Usage (1/1), Commonsense Reasoning (1/1)
- Tied in:Agent Level Benchmark
On average across the 15 shared benchmarks, GPT-5.2 scores 0.29 higher.
Largest single-benchmark gap: GDPval-AA — GPT-5.2 70.90 vs Gemini 3.0 Pro (Preview 11-2025) 35 (+35.90).
Page generated from structured model, pricing and benchmark records. No real-time LLM is used to write the prose.