GoogleGemini 2.5 Deep Think
Gemini 2.5 Deep Think is a Google proprietary reasoning model. It is not yet ranked: 1 independent result so far, and the index needs at least five across two categories. Last measured 20 May 2025.stale: no new result in six months
- Blended price
- –
- Output speed
- –
- First answer
- –
- Context
- –
- Full answer
- –
- Cost per run
- –
- Released
- –
- date not published by our sources
- Multimodal73
Versions
Google has shipped 2 models under this name. Each is ranked on its own results; a newer version often has fewer results so far, which holds its index nearer the average until more arrive.
| Model | Released | Index | Rank |
|---|---|---|---|
| Gemini 3 Deep Thinkthinking | 5 Feb 2026 | – | – |
| Gemini 2.5 Deep Thinkthinkingthis page | – | – | – |
Benchmark results
See all LLM benchmarks. One column per reasoning effort. Rank is among every configuration of every model on that benchmark. Hover a score for the run it came from.
Multimodal
| Benchmark | thinking | Source |
|---|---|---|
| MMMU (validation) | 84.0%#2 | MMMU |
Where it wins
Benchmarks where this configuration ranks in the top five of every configuration measured.
More from Google
See also
Questions people ask
- Why does Gemini 2.5 Deep Think have no BenchLeader Index score?
- The index needs at least five independent results across at least two categories; as of 25 Oct 2026 Gemini 2.5 Deep Think has 1 that count. Launch-post scores are never used, so it is ranked once independent publishers have measured it.
Data as of 25 Oct 2026. Compare with another model.
Cite as: BenchLeader, “Gemini 2.5 Deep Think: benchmarks, pricing, speed and rank”, data as of 25 Oct 2026, https://www.benchleader.com/models/gemini-2-5-deep-think.