Gemini 3 Pro Preview (low)
GoogleGeminiProprietary
Description
Gemini 3 Pro is the first model in the new Gemini 3 series. It is best for complex tasks that require broad world knowledge and advanced reasoning across modalities. Gemini 3 Pro uses dynamic thinking by default to reason through prompts, and features a 1 million-token input context window with 64k output tokens.
Release Date
2025-11-18
Parameters
—
Context Length
1.0M
Modalities
audio, image, pdf, text, video
Capability Radar
49
general
78
coding
87
reasoning
63
scienceest.
70
agents
80
multimodal
Science uses a reasoning proxy when dedicated science benchmarks are unavailable.
Rankings
| Domain | #Rank | Score | Source |
|---|---|---|---|
| Agentic Capability | 5 | 75.0 | LS |
| Code Ranking | 38 | 79.0 | AA |
| General Ranking | 94 | 66.0 | AA |
| Math Reasoning | 52 | 88.0 | AA |
| Multimodal Ranking | 79 | 42.0 | LS |
| Reasoning | 104 | 31.0 | LS |
| Science | 34 | 76.0 | AA |
Benchmark Scores (LLM Stats)
Agents
Vending-Bench 2
547816.0%SR
t2-bench
85.4%SR
Terminal-Bench 2.0
54.2%SR
Biology
GPQA
91.9%SR
Code
LiveCodeBench Pro
2439.00 / 3000SR
SWE-Bench Verified
76.2%SR
Factuality
SimpleQA
72.1%SR
FACTS Grounding
70.5%SR
General
Global PIQA
93.4%SR
MMMLU
91.8%SR
MMMU-Pro
81.0%SR
LiveBench
73.4%SR
MRCR v2 (8-needle)
26.3%SR
Grounding
ScreenSpot Pro
72.7%SR
Healthcare
VideoMMMU
87.6%SR
Math
AIME 2025
100.0%SR
Humanity's Last Exam
45.8%SR
MathArena Apex
23.4%SR
Multimodal
CharXiv-R
81.4%SR
OmniDocBench 1.5
11.5%SR
Reasoning
ARC-AGI v2
31.1%SR
AA Evaluation Indices
Math Index86.7
Intelligence Index33.1
Mmlu Pro0.9
Gpqa0.9
Aime 250.9
Livecodebench0.9
Tau20.7
Lcr0.7
Scicode0.5
Ifbench0.5
Terminalbench Hard0.3
Hle0.3
LLM Stats Category Scores
Agents100
Code100
Reasoning100
General100
Language90
Physics90
Healthcare90
Biology90
Chemistry90
Frontend Development80
Math70
Multimodal70
Factuality70
Grounding70
Tool Calling70
Vision60
Spatial Reasoning50
Long Context30
Structured Output10
Pricing
Input Price$2 / 1M tokens
Output Price$12 / 1M tokens
Blended Price (3:1)$4.5 / 1M tokens
Cache Read Price$0.2 / 1M tokens
Speed
Tokens/sec0.0
Time to First Token0.00s
Time to Answer0.00s
Provider Price Ranking
Provider Price Ranking
5 providers
Cheapest: QiHangMost Expensive: Helicone
ProviderInputOutput
1QiHangCheapest
$0.57
$3.43
2Jiekou.AI
$1.8
$10.8
3GooglePRIMARY
$2
$12
4302.AI
$2
$12
5Helicone
$2
$12
Compare pricing across different API providers for this model.