Gemini 3 Deep Think (2/26): benchmark scores, price and ranking

Gemini 3 Deep Think (2/26) from Google ranks #68 of 743 LLMs on the LLMs Tiger aggregate benchmark score (52.4), based on 1 benchmark results captured up to 2026-10-13.

Organisation: Google
Type: Frontier model
Open weights: No
Aggregate rank: #68 of 743 (score 52.4)

Across the 1 benchmark where at least ten models are scored, Gemini 3 Deep Think (2/26) is in the top 10% on 0 and in the top half on 1.

Gemini 3 Deep Think (2/26) benchmark results

Gemini 3 Deep Think (2/26) benchmark scores, ranked against all models measured on each benchmark
Benchmark	Score	Rank	Source	Captured
ARC-AGI-2	84.6%	#12 of 87	ARC Prize official leaderboard	2026-10-13

About Gemini 3 Deep Think (2/26)

Gemini 3 Deep Think (2/26) is a frontier model 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 (2/26) stands

On our aggregate score, Gemini 3 Deep Think (2/26) is #68 of 743 ranked models, in the top quarter. Its score is 52.4; the scale is centred on 50, so it sits 2.4 points above the typical model. That beats 91% of the ranked field. Among the 29 ranked Google models it is #6. The next one up is Gemini 2.5 Pro Preview 03-25 at 52.6. The next one down is gemini-3.5-flash-high at 52.2.

Gemini 3 Deep Think (2/26), benchmark by benchmark

ARC-AGI-2 puts Gemini 3 Deep Think (2/26) at 84.6%, #12 of 87. That beats 86% of the models tested, but 10.4 points short of GPT-6 Astra (max) (95.0%). The results either side are Gemini 3.7 Flash (high) (84.6%) and GPT-5.6 Terra (max) (83.9%). It covers abstract reasoning, ARC Prize public eval set. The board median is 22.8%, so Gemini 3 Deep Think (2/26) is 61.8 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

#65 Tencent Hy4 preview (unknown) (52.4)
#66 Muse Spark 1.1 (xhigh) (52.4)
#67 DeepSeek-V3.2-Exp (high) (52.4)
#69 AI co-mathematician (52.3)
#70 Darwin-180B-RSI (52.3)
#71 Qwen3.7 Max (52.3)