Google model product
Gemini 3 Pro is the first model in the new Gemini 3 series.
Updated Aug 17, 2026. Default version: Gemini 3 Pro
This profile uses the model's current scored version. Arena ratings and prices are shown separately.
Benchmark scores for Gemini 3 Pro.
Benchmark | Score | Rank | Participants | Percentile | Evidence | Evaluated |
|---|
| BenchmarkAIME 2025 | Score100.0% | Rank01 | Participants115 | Percentile100.0% | EvidenceC | Evaluated |
| BenchmarkGlobal PIQA | Score93.4% | Rank01 | Participants13 | Percentile100.0% | EvidenceC | Evaluated |
| BenchmarkVideoMMMU | Score87.6% | Rank01 | Participants26 | Percentile100.0% | EvidenceC | Evaluated |
| BenchmarkLiveCodeBench Pro | Score2,439 points | Rank02 | Participants4 | Percentile66.7% | EvidenceC | Evaluated |
| BenchmarkVending-Bench 2 | Score5,478.16 usd | Rank03 | Participants4 | Percentile33.3% | EvidenceC | Evaluated |
| BenchmarkMMMLU | Score91.8% | Rank04 | Participants49 | Percentile93.8% | EvidenceC | Evaluated |
| BenchmarkSimpleQA | Score72.1% | Rank06 | Participants47 | Percentile89.1% | EvidenceC | Evaluated |
| BenchmarkScreenSpot Pro | Score72.7% | Rank07 | Participants25 | Percentile75.0% | EvidenceC | Evaluated |
| Benchmarkt2-bench | Score85.4% | Rank07 | Participants23 | Percentile72.7% | EvidenceC | Evaluated |
| BenchmarkFACTS Grounding | Score70.5% | Rank08 | Participants13 | Percentile41.7% | EvidenceC | Evaluated |
| BenchmarkMathArena Apex | Score23.4% | Rank08 | Participants8 | Percentile0.0% | EvidenceC | Evaluated |
| BenchmarkMMMU-Pro | Score81.0% | Rank10 | Participants68 | Percentile86.6% | EvidenceC | Evaluated |
| BenchmarkARC-AGI v2 | Score31.1% | Rank13 | Participants17 | Percentile25.0% | EvidenceC | Evaluated |
| BenchmarkGPQA | Score91.9% | Rank16 | Participants239 | Percentile93.7% | EvidenceC | Evaluated |
| BenchmarkMRCR v2 (8-needle) | Score26.3% | Rank16 | Participants23 | Percentile31.8% | EvidenceC | Evaluated |
| BenchmarkOmniDocBench 1.5 | Score11.5% | Rank18 | Participants18 | Percentile0.0% | EvidenceC | Evaluated |
| BenchmarkCharXiv-R | Score81.4% | Rank19 | Participants51 | Percentile64.0% | EvidenceC | Evaluated |
| BenchmarkLiveBench | Score73.4% | Rank21 | Participants38 | Percentile46.0% | EvidenceB | Evaluated |
| BenchmarkHumanity's Last Exam | Score45.8% | Rank28 | Participants99 | Percentile72.5% | EvidenceC | Evaluated |
| BenchmarkTerminal-Bench 2.0 | Score54.2% | Rank31 | Participants51 | Percentile40.0% | EvidenceC | Evaluated |
| BenchmarkSWE-Bench Verified | Score76.2% | Rank35 | Participants111 | Percentile69.1% | EvidenceC | Evaluated |
Preference and agent-evaluation results for the default version.
Arena | Category | Rank | Rating / score | Votes | Observations | Result date |
|---|
| Arenasearch style control | Categoryoverall | Rank07 | Rating / score1201.2 | Votes37,255 | ObservationsN/A | Result date |
| Arenasearch | Categoryoverall | Rank08 | Rating / score1207.0 | Votes37,255 | ObservationsN/A | Result date |
| Arenatext factuality | Categoryoverall | Rank10 | Rating / score1480.7 | Votes40,615 | ObservationsN/A | Result date |
| Arenavision style control | Categoryoverall | Rank11 | Rating / score1289.1 | Votes13,095 | ObservationsN/A | Result date |
| Arenavision | Categoryoverall | Rank12 | Rating / score1304.9 | Votes13,095 | ObservationsN/A | Result date |
| Arenatext | Categoryoverall | Rank13 | Rating / score1479.3 | Votes41,505 | ObservationsN/A | Result date |
| Arenatext style control | Categoryoverall | Rank14 | Rating / score1485.5 | Votes41,505 | ObservationsN/A | Result date |
| Arenasearch factuality | Categoryoverall | Rank15 | Rating / score1193.7 | Votes37,134 | ObservationsN/A | Result date |
| Arenadocument style control | Categoryoverall | Rank24 | Rating / score1450.5 | Votes10,739 | ObservationsN/A | Result date |
| Arenadocument | Categoryoverall | Rank27 | Rating / score1433.0 | Votes10,739 | ObservationsN/A | Result date |
| Arenawebdev | Categoryoverall | Rank51 | Rating / score1437.8 | Votes13,585 | ObservationsN/A | Result date |
Official vendor API pricing appears first, followed by individual provider offers.
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 |
| ProviderOrcaRouter | Provider model IDgoogle/gemini-3-pro-preview | Regionglobal | Input / 1M$4 | Output / 1M$18 | Context1M | Updated |
Official prices use only the vendor's configured official Provider and positive standard USD PAYG rates. Third-party offers remain explicitly labeled.
Provider-specific output speed and catalog latency for Gemini 3 Pro. Runtime does not affect the capability score.
Provider | Output Speed | Catalog Latency | Max Input | Max Output | Updated |
|---|
| ProviderGoogle | Output Speed90 tok/s | Catalog Latency0.6 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.
Technical details for the model's default version.
Available versions of this model. The score column identifies the version used in the overall ranking.
Version | Released | LLMBoard | Parameters | Context | Max output | Open weights | License |
|---|
| VersionGemini 3 Pro | Released | LLMBoard67.1 | ParametersN/A | Context1M | Max output65.5K | Open weightsNo | LicenseProprietary |
Open a comparison with the three ranked models immediately above and below this model.
Recommendations prioritize the same model type and family, then the closest LLMBoard score.
Key information about Gemini 3 Pro and its available data.
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.
Data as of 2026-08-17.
Common questions about Gemini 3 Pro.
Gemini 3 Pro's default version was released on Nov 18, 2025.
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.
Gemini 3 Pro was created by Google.
The default version has a 1M token context window.
No. The default version is not marked as having publicly available weights.
9 provider offerings are linked to the default version.
Nearby ranked alternatives include GLM 5.1, Seed 2.0 Pro, Grok 4 Heavy.