- Services
Gemini is Google’s closed-source frontier model family. Gemini 2.5 Pro and Flash power Vertex AI, the Gemini app, and Google Search’s AI Overviews. Variants span fast multimodal models to long-context reasoning tiers.
See all models from GoogleModels in family
22
Open weight
0
API only
22
Avg score
64.4
Top benchmark
100.0
Video Arena
Total HF downloads
—
Primary modality
Text
Context window
131K – 1.0M
First release
Feb 2025
Latest release
Nov 2026
Every release in the Gemini family, ranked by composite score across benchmarks, popularity, efficiency, and versatility.
| # | Model | Modality | Score | Params | Released |
|---|---|---|---|---|---|
| 1 | Gemini Omni FlashAPI | video | SS86.9 | — | May 2026 |
| 2 | Gemini 3 FlashAPI | text | AA75.5 | — | Dec 2025 |
| 3 | Gemini 3 Pro Image (Nano Banana Pro)API | image | AA73.8 | — | Nov 2025 |
| 4 | Gemini 3 ProAPI | text | AA71.7 | — | Nov 2025 |
| 5 | Gemini 2.0 FlashAPI | text | BB69.3 | — | Feb 2025 |
| 6 | Gemini 3.5 FlashAPI | text | BB68.2 | — | May 2026 |
| 7 | Gemini 2.5 FlashAPI | text | BB68.0 | — | May 2025 |
| 8 | Gemini 3.1 Pro PreviewAPI | text | BB68.0 | — | Feb 2026 |
| 9 | Gemini 3 Flash (Thinking Minimal)API | text | BB67.4 | — | Dec 2025 |
| 10 | Gemini 2.5 ProAPI | text | BB67.0 | — | Mar 2025 |
When a family spans multiple modalities, the strongest model in each one.
Text
Gemini 3 Flash
Image
Gemini 3 Pro Image (Nano Banana Pro)
Video
Gemini Omni Flash
Audio
Gemini 3.8 Flash TTS
How the family splits across text, image, video, audio, embedding, and decision releases.
When each release shipped, newest first. Useful for tracking version cadence.
Nov 5
Oct 28
Sep 30
Sep 23
May 19
May 19
May 19
Mar 2
Feb 26
Feb 18
Dec 10
Dec 10
Nov 20
Nov 18
Sep 24
Sep 24
Aug 26
Jun 16
May 19
Mar 24
Feb 4
Feb 4
Composite grades across this family. Higher is better, blending benchmarks, popularity, and efficiency.
Closed-source releases accessed through Google’s API.
| # | Model | Modality | Score | Params | Released |
|---|---|---|---|---|---|
| 1 | Gemini Omni FlashAPI | video | SS86.9 | — | May 2026 |
| 2 | Gemini 3 FlashAPI | text | AA75.5 | — | Dec 2025 |
| 3 | Gemini 3 Pro Image (Nano Banana Pro)API | image | AA73.8 | — | Nov 2025 |
| 4 | Gemini 3 ProAPI | text | AA71.7 | — | Nov 2025 |
| 5 | Gemini 2.0 FlashAPI | text | BB69.3 | — | Feb 2025 |
| 6 | Gemini 3.5 FlashAPI | text | BB68.2 | — | May 2026 |
| 7 | Gemini 2.5 FlashAPI | text | BB68.0 | — | May 2025 |
| 8 | Gemini 3.1 Pro PreviewAPI | text | BB68.0 | — | Feb 2026 |
| 9 | Gemini 3 Flash (Thinking Minimal)API | text | BB67.4 | — | Dec 2025 |
| 10 | Gemini 2.5 ProAPI | text | BB67.0 | — | Mar 2025 |
| 11 | Gemini 3.8 Flash TTSAPI | audio | BB65.0 | — | Sep 2026 |
| 12 | Gemini 4 CarbonAPI | text | BB64.4 | — | Oct 2026 |
| 13 | Gemini 2.5 Flash Lite Preview 06-17 (Thinking)API | text | BB63.1 | — | Jun 2025 |
| 14 | Gemini 2.5 Flash Image (Nano Banana)API | image | BB60.9 | — | Aug 2025 |
| 15 | Gemini 2.5 Flash Lite Preview 09-2025 (No Thinking)API | text | BB58.9 | — | Sep 2025 |
| 16 | Gemini 2.5 Flash Preview 09-2025API | text | BB58.3 | — | Sep 2025 |
| 17 | Gemini 2.0 Flash Lite PreviewAPI | text | BB57.9 | — | Feb 2025 |
| 18 | Gemini 3.1 Flash Lite PreviewAPI | text | BB57.9 | — | Mar 2026 |
| 19 | Gemini 4 ArgonAPI | text | BB57.3 | — | Sep 2026 |
| 20 | Gemini OmniAPI | video | BB57.0 | — | May 2026 |
| 21 | Gemini Omni 1.2 FlashAPI | video | CC51.7 | — | Nov 2026 |
| 22 | Gemini 3.1 Flash ImageAPI | image | CC48.8 | — | Feb 2026 |
The Gemini family is a series of AI models from Google. This page lists every release in the family with its benchmark scores, parameter count, and hardware requirements.
By composite score, Gemini Omni Flash is currently the top model in the family. For local inference, match the parameter count to your VRAM budget. For quality, pick the highest scorer that fits.
See the open-weight section above for models you can run locally. The API-only section lists closed releases that must be accessed through the provider’s API.
Spin up an instance in the cloud, or pick local hardware that fits.
Advertising disclosure: we earn commissions when you shop through the links below.
Vast.ai
Decentralized GPU marketplace with the lowest hourly prices.
RunPod
Pay-per-second GPU rentals starting at $0.20/hr.
Digital Ocean
Spin up a GPU droplet in minutes, starting at $0.75/hr.
Vultr
GPU cloud with hourly and monthly plans, starting at $0.50/hr.
GPU Mart
Dedicated GPU servers and VPS billed monthly, starting at $0.50/hr.
PPQ.ai
Multi-model inference gateway for production workloads.
Find Local Hardware
See which GPU, Mac, or workstation can run Gemini on-prem.
Local LLM Mini PCs
Compact machines that run Gemini on your desk. Chosen for local inference.
Benchmark data sourced from: Hugging Face, Arena.ai, Artificial Analysis.
