Google model product
Gemini 3 Pro
Gemini 3 Pro is a Google model for complex tasks requiring broad world knowledge and advanced reasoning across modalities.
Updated Sep 8, 2026. Default version: Gemini 3 Pro
LLMBoard Score66.1Gemini 3 Pro
Coverage40%21 benchmark families
Context window1MTokens
Official input priceN/AOfficial price unavailable
Gemini 3 Pro Capability Profile
This profile uses the model's current scored version. Arena ratings and prices are shown separately.
Gemini 3 Pro Benchmark Results
Benchmark scores for Gemini 3 Pro.
Sort by
Benchmark | Score | Rank | Participants | Percentile | Evidence | Evaluated |
|---|
| BenchmarkAIME 2025 | Score100.00% | Rank01 | Participants119 | Percentile100.00% | EvidenceC | Evaluated |
| BenchmarkGlobal PIQA | Score93.40% | Rank01 | Participants13 | Percentile100.00% | EvidenceC | Evaluated |
| BenchmarkVideoMMMU | Score87.60% | Rank01 | Participants26 | Percentile100.00% | EvidenceC | Evaluated |
| BenchmarkLiveCodeBench Pro | Score2,439.00 points | Rank02 | Participants4 | Percentile66.67% | EvidenceC | Evaluated |
| BenchmarkMMMLU | Score91.80% | Rank03 | Participants49 | Percentile95.83% | EvidenceC | Evaluated |
| BenchmarkVending-Bench 2 | Score5,478.16 usd | Rank03 | Participants4 | Percentile33.33% | EvidenceC | Evaluated |
| BenchmarkSimpleQA | Score72.10% | Rank06 | Participants47 | Percentile89.13% | EvidenceC | Evaluated |
| Benchmarkt2-bench | Score85.40% | Rank07 | Participants23 | Percentile72.73% | EvidenceC | Evaluated |
| BenchmarkFACTS Grounding | Score70.50% | Rank08 | Participants13 | Percentile41.67% | EvidenceC | Evaluated |
| BenchmarkScreenSpot Pro | Score72.70% | Rank08 | Participants26 | Percentile72.00% | EvidenceC | Evaluated |
| BenchmarkLM Arena Vision | Score1,304.45 rating | Rank08 | Participants107 | Percentile93.40% | EvidenceA | Evaluated |
| BenchmarkLM Arena Vision Style Control | Score1,288.59 rating | Rank08 | Participants107 | Percentile93.40% | EvidenceA | Evaluated |
| BenchmarkMathArena Apex | Score23.40% | Rank09 | Participants9 | Percentile0.00% | EvidenceC | Evaluated |
| BenchmarkLM Arena Search | Score1,207.35 rating | Rank09 | Participants28 | Percentile70.37% | EvidenceA | Evaluated |
| BenchmarkLM Arena Search Style Control | Score1,201.38 rating | Rank09 | Participants28 | Percentile70.37% | EvidenceA | Evaluated |
| BenchmarkMMMU-Pro | Score81.00% | Rank10 | Participants69 | Percentile86.76% | EvidenceC | Evaluated |
| BenchmarkLM Arena Text Factuality | Score1,481.03 rating | Rank10 | Participants121 | Percentile92.50% | EvidenceA | Evaluated |
| BenchmarkLM Arena Text | Score1,479.48 rating | Rank12 | Participants210 | Percentile94.74% | EvidenceA | Evaluated |
| BenchmarkLM Arena Text Style Control | Score1,485.58 rating | Rank13 | Participants210 | Percentile94.26% | EvidenceA | Evaluated |
| BenchmarkARC-AGI v2 | Score31.10% | Rank14 | Participants18 | Percentile23.53% | EvidenceC | Evaluated |
| BenchmarkLM Arena Search Factuality | Score1,193.86 rating | Rank14 | Participants28 | Percentile51.85% | EvidenceA | Evaluated |
| BenchmarkGPQA | Score91.90% | Rank18 | Participants247 | Percentile93.09% | EvidenceC | Evaluated |
| BenchmarkMRCR v2 (8-needle) | Score26.30% | Rank18 | Participants24 | Percentile26.09% | EvidenceC | Evaluated |
| BenchmarkOmniDocBench 1.5 | Score11.50% | Rank18 | Participants18 | Percentile0.00% | EvidenceC | Evaluated |
| BenchmarkLM Arena Document Style Control | Score1,450.50 rating | Rank20 | Participants33 | Percentile40.63% | EvidenceA | Evaluated |
| BenchmarkLiveBench | Score73.39% | Rank21 | Participants38 | Percentile45.95% | EvidenceB | Evaluated |
| BenchmarkCharXiv-R | Score81.40% | Rank23 | Participants55 | Percentile59.26% | EvidenceC | Evaluated |
| BenchmarkLM Arena Document | Score1,433.01 rating | Rank23 | Participants33 | Percentile31.25% | EvidenceA | Evaluated |
| BenchmarkHumanity's Last Exam | Score45.80% | Rank30 | Participants103 | Percentile71.57% | EvidenceC | Evaluated |
| BenchmarkTerminal-Bench 2.0 | Score54.20% | Rank31 | Participants51 | Percentile40.00% | EvidenceC | Evaluated |
Gemini 3 Pro Arena Results
Preference and agent-evaluation results for the default version.
Sort by
Arena | Category | Rank | Rating / score | Votes | Observations | Result date |
|---|
| Arenatext factuality | Categorygerman | Rank01 | Rating / score1,479.64 | Votes383 | ObservationsN/A | Result date |
| Arenatext factuality | Categorypolish | Rank01 | Rating / score1,487.61 | Votes866 | ObservationsN/A | Result date |
| Arenatext style control | Categorygerman | Rank01 | Rating / score1,521.60 | Votes797 | ObservationsN/A | Result date |
| Arenavision | Categorycreative writing | Rank01 | Rating / score1,390.64 | Votes766 | ObservationsN/A | Result date |
| Arenavision | Categoryentity recognition | Rank01 | Rating / score1,353.92 | Votes242 | ObservationsN/A | Result date |
| Arenavision style control | Categorycreative writing | Rank01 | Rating / score1,372.90 | Votes766 | ObservationsN/A | Result date |
| Arenavision style control | Categorycreative writing vision | Rank01 | Rating / score1,323.59 | Votes1,030 | ObservationsN/A | Result date |
| Arenavision style control | Categoryentity recognition | Rank01 | Rating / score1,298.19 | Votes242 | ObservationsN/A | Result date |
| Arenatext | Categorygerman | Rank02 | Rating / score1,516.63 | Votes797 | ObservationsN/A | Result date |
| Arenatext | Categoryjapanese | Rank02 | Rating / score1,510.60 | Votes456 | ObservationsN/A | Result date |
| Arenatext factuality | Categoryindustry medicine and healthcare | Rank02 | Rating / score1,521.10 | Votes1,416 | ObservationsN/A | Result date |
| Arenatext factuality | Categoryspanish | Rank02 | Rating / score1,488.70 | Votes754 | ObservationsN/A | Result date |
| Arenavision | Categorycaptioning | Rank02 | Rating / score1,313.70 | Votes216 | ObservationsN/A | Result date |
| Arenavision | Categorycreative writing vision | Rank02 | Rating / score1,340.11 | Votes1,030 | ObservationsN/A | Result date |
| Arenavision style control | Categorycaptioning | Rank02 | Rating / score1,266.16 | Votes216 | ObservationsN/A | Result date |
| Arenatext | Categorypolish | Rank03 | Rating / score1,503.09 | Votes1,332 | ObservationsN/A | Result date |
| Arenatext style control | Categorypolish | Rank03 | Rating / score1,505.08 | Votes1,332 | ObservationsN/A | Result date |
| Arenavision | Categoryhumor | Rank03 | Rating / score1,339.32 | Votes544 | ObservationsN/A | Result date |
| Arenavision style control | Categoryhumor | Rank03 | Rating / score1,317.31 | Votes544 | ObservationsN/A | Result date |
| Arenatext factuality | Categorychinese | Rank04 | Rating / score1,537.19 | Votes1,677 | ObservationsN/A | Result date |
| Arenatext factuality | Categorycreative writing | Rank05 | Rating / score1,477.54 | Votes3,733 | ObservationsN/A | Result date |
| Arenatext style control | Categoryjapanese | Rank05 | Rating / score1,505.34 | Votes456 | ObservationsN/A | Result date |
| Arenavision | Categorychinese | Rank05 | Rating / score1,370.39 | Votes1,517 | ObservationsN/A | Result date |
| Arenatext factuality | Categorykorean | Rank06 | Rating / score1,451.88 | Votes330 | ObservationsN/A | Result date |
| Arenavision style control | Categorychinese | Rank06 | Rating / score1,344.58 | Votes1,517 | ObservationsN/A | Result date |
| Arenatext style control | Categorycreative writing | Rank07 | Rating / score1,483.10 | Votes6,236 | ObservationsN/A | Result date |
| Arenatext style control | Categoryindustry entertainment and sports and media | Rank07 | Rating / score1,467.71 | Votes7,595 | ObservationsN/A | Result date |
| Arenavision style control | Categoryocr | Rank07 | Rating / score1,301.85 | Votes7,957 | ObservationsN/A | Result date |
| Arenatext | Categorycreative writing | Rank08 | Rating / score1,480.55 | Votes6,236 | ObservationsN/A | Result date |
| Arenatext | Categoryindustry entertainment and sports and media | Rank08 | Rating / score1,463.09 | Votes7,595 | ObservationsN/A | Result date |
Gemini 3 Pro Pricing
Official vendor API pricing appears first, followed by individual provider offers.
- Official API
- N/A
- Official provider
- N/A
- Lowest third-party
- From $0.57 input, $3.43 output per 1M via QiHang
- Tracked offerings
- 8
Sort by
Provider | Provider model ID | Region | Input / 1M | Output / 1M | Context | Updated |
|---|
| ProviderQiHang | Provider model IDgemini-3-pro-preview | Regionglobal | Input / 1M$0.57 | Output / 1M$3.43 | Context1M | Updated |
| ProviderPoe | Provider model IDgoogle/gemini-3-pro | Regionglobal | Input / 1M$1.6 | Output / 1M$9.6 | Context1M | Updated |
| ProviderJiekou.AI | Provider model IDgemini-3-pro-preview | Regionglobal | Input / 1M$1.8 | Output / 1M$10.8 | Context1M | Updated |
| ProviderMerge Gateway | Provider model IDgoogle/gemini-3-pro-preview | Regionglobal | Input / 1M$2 | Output / 1M$12 | Context1M | Updated |
| ProviderHelicone | Provider model IDgemini-3-pro-preview | Regionglobal | Input / 1M$2 | Output / 1M$12 | Context1M | Updated |
| Provider302.AI | Provider model IDgemini-3-pro-preview | Regionglobal | Input / 1M$2 | Output / 1M$12 | Context1M | Updated |
| ProviderOpenCode Zen | Provider model IDgemini-3-pro | Regionglobal | Input / 1M$2 | Output / 1M$12 | Context1M | Updated |
Official prices use only the vendor's configured official Provider and positive standard USD PAYG rates. Third-party offers remain explicitly labeled.
Gemini 3 Pro Runtime Performance
Provider-specific output speed and catalog latency for Gemini 3 Pro. Runtime does not affect the capability score.
Sort by
Provider | Output Speed | Catalog Latency | Max Input | Max Output | Updated |
|---|
| ProviderGoogle | Output Speed90.00 tok/s | Catalog Latency0.60 s | Max Input1M | Max Output65.5K | Updated |
Output Speed is generated output tokens received per second. Catalog latency is reported separately from observed provider TTFT.
Gemini 3 Pro Specifications
Technical details for the model's default version.
- Version
- Gemini 3 Pro
- Released
- Nov 18, 2025
- Knowledge cutoff
- Jan 31, 2025
- Parameters
- N/A
- Context window
- 1M
- Max output
- 65.5K
- Inputs
- audio, image, text, video
- Outputs
- text
- Open weights
- No
- License
- Proprietary
Gemini 3 Pro Versions
Available versions of this model. The score column identifies the version used in the overall ranking.
Sort by
Version | Released | LLMBoard | Parameters | Context | Max output | Open weights | License |
|---|
| VersionGemini 3 Pro | Released | LLMBoard66.14 | ParametersN/A | Context1M | Max output65.5K | Open weightsNo | LicenseProprietary |
Models similar to Gemini 3 Pro
Recommendations prioritize the same model type and family, then the closest LLMBoard score.
GO
Google
Gemini 3 Flash
GOGoogle
Gemini 3.5 Flash Cyber
GOGoogle
Gemini 3.8 Flash Cyber
GOGoogle
Gemma 4 31B
GOGoogle
Gemini 3.6 Flash
GOGoogle
Gemini 3.1 Pro
What is Gemini 3 Pro?
Key information about Gemini 3 Pro and its available data.
Gemini 3 Pro is the first model in the Gemini 3 series, with dynamic thinking enabled by default. It has a 1 million-token input context window and supports up to 64k output tokens.
Data as of 2026-09-08.
FAQ
Common questions about Gemini 3 Pro.
When was Gemini 3 Pro released?
Gemini 3 Pro's default version was released on Nov 18, 2025.
How much does Gemini 3 Pro cost?
No official standard PAYG price is currently available for Gemini 3 Pro. The lowest tracked third-party offer starts at $0.57 input and $3.43 output via QiHang.
Who created Gemini 3 Pro?
Gemini 3 Pro was created by Google.
What is the context window for Gemini 3 Pro?
The default version has a 1M token context window.
Is Gemini 3 Pro open weight?
No. The default version is not marked as having publicly available weights.
How many API providers offer Gemini 3 Pro?
8 provider offerings are linked to the default version.