Compare Models
Gemini 3 Pro (11/25)

Release Date: Nov 18, 2025

Developer Google 🇺🇸
Context Window 1M
Max Output Tokens 66k
Token Costs (in/out) $2.00/12.00
Weights Private
Input Modalities

Accuracy

69.38 %

Avg. Cost (In/Out)

$ 2.00 / $ 12.00

Latency

14 min 59 s

Vals Index
BenchmarksAccuracyRankings
MedCode

0.0%

±2.07
12/92
MedScribe

0.0%

±1.90
77/94
MortgageTax

0.0%

±0.91
7/98
SAGE

0.0%

±3.38
28/81
TaxEval v2

0.0%

±0.87
62/145
Vibe Code Bench v1.1

0.0%

±3.06
72/95
GPQA Diamond

0.0%

±1.39
19/138
LiveCodeBench

0.0%

±0.98
23/143
LegalBench

0.0%

±0.37
5/144
MMLU Pro

0.0%

±0.29
7/138
MMMU Pro

0.0%

±0.80
14/93
SWE-bench

0.0%

±1.90
39/88
Terminal-Bench 2.0

0.0%

±5.30
15/67
Proprietary BenchmarksAcademic BenchmarksIndustry Partners
Vals
Default Provider : Google
Temperature: 1
Top P: Default
Top K: Default
Max Output Tokens: 65,536
Reasoning Effort: high

Updates

Nov 18, 2025

We just evaluated Gemini 3 Pro (11/25) on our Vals Index! Key takeaways:

Gemini 3 Pro (11/25) demonstrates strong performance across the board, placing first on SAGE, GPQA Diamond and MortgageTax and second on our Multimodal Vals Index.

Gemini 3 Pro (11/25) is three times faster than GPT 5.1, and is also cheaper than Claude Sonnet 4.5 (Thinking), the leader on the Vals Multimodal Index.

Overall, Gemini 3 Pro (11/25) excels in multimodal use cases and provides meaningful improvement over its predecessor, Gemini 2.5 Pro Exp.