이 모델의 강점은 무엇인가요?

고급 추론 능력 Google DeepMind 제작 논리적 사고 특화

이 모델의 약점은 무엇인가요?

소스 코드 비공개 상업적 사용 제한 외부 당사자에 대한 투명성 부족

어떤 용도에 가장 적합한가요?

복잡한 문제 해결 논리적 분석 작업 고급 추론을 요구하는 작업

모델 목록으로

Google Deep Mind독점

Gemini 3 Pro – Deep Think Mode

Name: Gemini 3 Pro – Deep Think Mode
Author: Google Deep Mind

Gemini 3 Pro – Deep Think Mode는 Google DeepMind가 개발한 추론 모델입니다. 고급 논리적 사고에 특화되도록 설계되었으며, 폐쇄형 라이선스로 제공됩니다.

파라미터

Undisclosed

컨텍스트

라이선스

Proprietary

출시일

2025-12-04

API 가격

이 모델의 API 가격 정보는 현재 공개되지 않았습니다

강점

・고급 추론 능력
・Google DeepMind 제작
・논리적 사고 특화

약점

・소스 코드 비공개
・상업적 사용 제한
・외부 당사자에 대한 투명성 부족

활용 사례

・복잡한 문제 해결
・논리적 분석 작업
・고급 추론을 요구하는 작업

심층 분석

Arena Elo (Overall)

1486

Based on 41,631 votes on Chatbot Arena

Arena Elo (Coding)

1519

Top coding Elo, 8,622 votes

Arena Elo (Hard Prompts)

1504

22,573 votes

ARC-AGI-2

84.6%

ARC Prize verified, #1 among all models

Context Window

2M tokens

Largest context in the family (BenchLM reports 2M)

Overall Score (BenchLM)

90/100

#7 of 117 models on provisional leaderboard

Agentic Rank

Score 95.4/100, strong on Terminal-Bench, BrowseComp, OSWorld

Multimodal Rank

Score 100/100, strongest on MMMU-Pro, CharXiv

Release Date

February 12, 2026

Same release as Gemini 3 Deep Think upgrade

강점

・Arena Elo 1486 overall, 1519 in coding - top-tier human preference rankings
・Best multimodal model (#1 rank, 100/100 score) for screenshots, documents, charts
・2M token context window (largest in the Gemini 3 family)
・Strong agentic performance (#4 rank, 95.4/100) on tool use and computer tasks
・Excellent hard prompt handling (Elo 1504)

약점

・Coding benchmark rank is #30 (74.5/100) despite high Arena coding Elo - discrepancy between benchmarks and human preference
・Only 4 of 221 tracked benchmarks have published scores (limited verified data)
・Higher cost than standard Gemini 3 Pro ($2/$12 per 1M tokens)
・Deep Think mode increases latency significantly
・Not yet broadly available via API (early access program)

경쟁사 비교

Model	Arena	SWE	GPQA	Price
Gemini 3 Pro Deep Think	1486	~76%	~94%	$2/$12 per 1M
Gemini 3.1 Pro	~1490	~78%	~92%	$2/$12 per 1M
Claude Opus 4.6	~1490	~81%	91.3%	$15/$75 per 1M
Gemini 2.5 Pro	1465	63.8%	84%	$1.25/$10 per 1M

개요

Gemini 3 Pro Deep Think Mode is the reasoning-enhanced variant of Gemini 3 Pro, achieving Arena Elo 1486 overall and 1519 in coding on Chatbot Arena. It ranks #1 in multimodal tasks and #4 in agentic benchmarks. Released February 12, 2026, it combines deep chain-of-thought reasoning with Gemini 3 Pro's base capabilities, excelling at complex multimodal workflows and hard prompts.

벤치마크 및 성능

Arena Elo 1486 (overall), 1519 (coding), 1504 (hard prompts). BenchLM provisional score 90/100 (#7 of 117). Strongest categories: Multimodal (#1, 100/100) and Agentic (#4, 95.4/100). Weakest: Coding benchmarks (#30, 74.5/100) despite high Arena coding Elo.

상세 비교

Competitive with Gemini 3.1 Pro at the same price point. Excels at multimodal understanding where it outperforms all competitors. For pure coding tasks, Claude Opus 4.6 leads. For abstract reasoning, the Deep Think mode gives it an edge over standard Gemini 3 Pro.

커뮤니티 평가

Well-regarded for complex multimodal tasks. Users note strong performance on document understanding, chart analysis, and visual reasoning. The 2M context window is appreciated for long-document workflows. Some users report latency issues with the deep reasoning mode.

활용 사례

Best for multimodal reasoning tasks combining text, images, charts, and documents. Ideal for visual question answering, document analysis, complex agentic workflows with tool use, and tasks requiring both visual understanding and deep logical reasoning.