Gemini 3 Deep Think (Preview) ²: benchmark scores, price and ranking
Gemini 3 Deep Think (Preview) ² from Google ranks #214 of 743 LLMs on the LLMs Tiger aggregate benchmark score (50.1), based on 1 benchmark results captured up to 2026-10-13.
- Organisation
- Type
- Preview release
- Open weights
- No
- Aggregate rank
- #214 of 743 (score 50.1)
Across the 1 benchmark where at least ten models are scored, Gemini 3 Deep Think (Preview) ² is in the top 10% on 0 and in the top half on 1.
Gemini 3 Deep Think (Preview) ² benchmark results
| Benchmark | Score | Rank | Source | Captured |
|---|---|---|---|---|
| ARC-AGI-2 | 45.1% | #38 of 87 | ARC Prize official leaderboard | 2026-10-13 |
About Gemini 3 Deep Think (Preview) ²
Gemini 3 Deep Think (Preview) ² is a preview release from Google. No API price is listed for it yet. We hold 1 benchmark result for it across 1 area, and none is in the top tenth of its board.
Where Gemini 3 Deep Think (Preview) ² stands
On our aggregate score, Gemini 3 Deep Think (Preview) ² is #214 of 743 ranked models, around the middle of the table. Its score is 50.1; the scale is centred on 50, so it sits 0.1 points above the typical model. That beats 71% of the ranked field. Among the 29 ranked Google models it is #10. The next one up is gemini-2.5-flash-preview-04-17 at 50.4. The next one down is gemini-exp-1114 at 50.0.
Gemini 3 Deep Think (Preview) ², benchmark by benchmark
ARC-AGI-2 puts Gemini 3 Deep Think (Preview) ² at 45.1%, #38 of 87. That beats 56% of the models tested, but 49.9 points short of GPT-6 Astra (max) (95.0%). The results either side are Grok 4.5 (high) (52.6%) and Qwen3.8-27B (XHigh) (42.4%). It covers abstract reasoning, ARC Prize public eval set. The board median is 22.8%, so Gemini 3 Deep Think (Preview) ² is 22.4 points above it. Source: ARC Prize official leaderboard, captured 2026-10-13.
Reading these numbers
The 1 score on this page comes from 1 source: ARC Prize official leaderboard (1). 1 of them come from an official leaderboard and 0 from datasets or third-party runs. With 1 results the aggregate rank can move a lot when one new score arrives.
Models ranked nearby
- #211 ollama/qwen2.5-coder:14b (50.1)
- #212 Qwen 3.6 Plus (2026-04-02) (50.1)
- #213 Qwen2.5-72B (50.1)
- #215 glm-4.7-flash (50.1)
- #216 grok-3-beta (50.0)
- #217 Claude Sonnet 4.6 (max) (50.0)
More from Google
- gemini-4-argon-high (#10, 59.5)
- gemini-3.1-pro-preview-high (#16, 55.3)
- gemini-2.5-pro-preview-06-05 (#35, 53.5)
- Gemini 2.5 Pro Preview 05-06 (#48, 53.0)
- Gemini 2.5 Pro Preview 03-25 (#59, 52.6)
- Gemini 3 Deep Think (2/26) (#68, 52.4)
Compare it on the live leaderboard → · How the aggregate score works