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。
Gemini 2.5 Flash
Google Deep Mind
Each axis is a category average, normalized to a 100-point radar.
Relative edge: none clear / Relative gap: 常识问答 -42.4
Relative edge: 常识问答 +42.4 / Relative gap: none clear
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
Gemini 3.0 Flash · 65.97
Best single
Gemini 3.0 Flash · AIME2025 99.70
Modality coverage
Gemini 2.5 Flash · 4 modalities
Head to head
12
Benchmarks
0
Wins
11
Losses
-19.55
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.
12 benchmarks with comparable scores. Each model shows its best score; mode label is displayed below.
| Benchmark | Gemini 2.5 Flash | Gemini 3.0 Flash |
|---|---|---|
HLE 综合评估 | 11.00Thinking Enabled | 43.50Thinking Enabled ｜ Tools |
LiveBench 综合评估 | 47.74Thinking Level · High | 72.40Thinking Level · High |
GPQA Diamond 科学与综合推理 | 82.80Thinking Enabled | 90.40Thinking Enabled |
SimpleQA 常识问答 | 26.90Thinking Enabled | 68.70Thinking Enabled |
SWE-bench Verified 编程与软件工程 | 50.00Standard Mode | 68.70Thinking Enabled |
WeirdML v2 编程与软件工程 | 40.9516K ｜ Tools | 61.60Standard Mode ｜ Tools |
AIME2025 数学推理 | 72.00Thinking Enabled | 99.70Thinking Enabled ｜ Tools |
4.20Standard Mode | 4.20Standard Mode | |
SimpleBench 常识推理 | 41.20Standard Mode | 61.10Standard Mode |
BALROG Agent能力评测 | 33.50Standard Mode ｜ Tools | 48.10Standard Mode ｜ Tools |
Pinch Bench OpenClaw智能体能力综合测评 | 70.70Thinking Enabled ｜ Tools | 85.20Thinking Enabled ｜ Tools |
GeoBench ACW 多模态理解 | 76.00Standard Mode | 88.00Standard Mode |
Side-by-side input/output token pricing
Licensing, MoE architecture, and multi-modality support.
| Features & specs | Gemini 2.5 FlashGoogle Deep Mind | Gemini 3.0 FlashGoogle Deep Mind |
|---|---|---|
Core specsRelease | 2025-04-17 | 2025-12-17 |
Context length | 1000K | 2000K |
Max output | 65536 | 65536 |
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 | / | / |
Audio Input/Output | / | / |
Video Input/Output | / | / |
ResourcesPaper / report | Start building with Gemini 2.5 Flash | Gemini 3 Flash: frontier intelligence built for speed |
DataLearner blog | 谷歌发布新一代大模型Gemini 2.5 Flash，成本、速度和性能的最优均衡，同时支持推理和非推理模式，评测结果超Sonnet 3.7 | Gemini 3 Flash：Google 在 12 月 17 日发布的新一代默认模型 |
Gemini 3.0 Flash
Google Deep Mind
