Artificial Analysis releases Gemini Omni Flash 1.2 video evaluation and leaderboards
11/11/2026 — 11/11, 08:18·1 sources (all)·8 reports (all)
Key takeaways
- Artificial Analysis has published a video evaluation and leaderboard for Gemini Omni Flash 1.2, covering AA-Video-T2V, AA-Video-I2V, and AA-…
- these figures come from Artificial Analysis reporting and estimates.
Background & analysis
On November 11, 2026, AI benchmarking platform Artificial Analysis announced a new Search API leaderboard via its official X account (@artificialanlys). Alongside the ranking, the organization published a methodology document that details the exact configuration used in Keenable Search, so that evaluation conditions are transparent and results can be checked externally. Artificial Analysis also stated that every Search API result in the leaderboard was produced by Stirrup, the team's open-source agent harness, and shared a link to Stirrup so others can reproduce the evaluations. This is currently the only available reporting on the release, with no vendor responses or follow-up developments recorded yet.
AI-generated from 1 reports · updated 3 days ago
Latest turnAccording to Artificial Analysis, Gemini Omni Flash 1.2 has been added to their video evaluation leaderboards. Users can check its performance across AA-Video-T2V, AA-Video-I2V, and AA-Video-Editing benchmarks, as well as vote in the Video Arena, providing a third-party reference for evaluating multimodal video generation and editing capabilities.
Related AI tools
- Heat index
- 25
- All sources
- 1
- All reports
- 8
- First seen
- 3 days ago
Reports on this story headlines open the original
According to Artificial Analysis, Gemini Omni Flash 1.2 has been added to their video evaluation leaderboards. Users can check its performance across AA-Video-T2V, AA-Video-I2V, and AA-Video-Editing benchmarks, as well as vote in the Video Arena, providing a third-party reference for evaluating multimodal video generation and editing capabilities.
X：Artificial Analysis (@artificialanlys)First-partyAI score 48According to Artificial Analysis, Gemini Omni Flash 1.2 ranks #1 in complex edits and visual effects editing on the AA-Video-Editing v1.1 leaderboard, outperforming Wan 3.0, while holding #2 in object editing. This demonstrates the model's updated capabilities in video post-processing and visual effects.
X：Artificial Analysis (@artificialanlys)First-partyAI score 54据 Artificial Analysis 报道，Gemini Omni Flash 1.2 在文本提示词生成视频（T2V）和图像提示词生成视频（I2V）中展现出不同特性，以起始帧输入时其在 20 项能力中较前代版本有显著提升。该评测揭示了多模态输入对视频生成质量的具体影响。
X：Artificial Analysis (@artificialanlys)First-partyAI score 52According to Artificial Analysis, Gemini Omni Flash 1.2 excels in dialogue scenes and designed content generation on the AA-Video-T2V leaderboard, rising to #1 in dialogue and lip sync, and #2 in multi-scene narrative. This reflects the model's capability evolution in specific video vertical use cases.
X：Artificial Analysis (@artificialanlys)First-partyAI score 52According to Artificial Analysis benchmarks, Gemini Omni Flash 1.2 generates a 10-second 1080p video in roughly 76 seconds, outperforming several higher-ranked video generation models tested. It outpaces competitors like MiniMax H3 and Wan 3.0 in rendering speed.
X：Artificial Analysis (@artificialanlys)First-partyAI score 52Gemini Omni Flash 1.2 costs $6.26 per minute of 1080p video, representing a 31% reduction compared to version 1.1, per Artificial Analysis. Google prices the model at $12.00 per million output video tokens on the Gemini API, down from $17.50.
X：Artificial Analysis (@artificialanlys)First-partyAI score 56Gemini Omni Flash 1.2, Google's video model released on November 5, secured the #2 spot on AA-Video-Editing v1.1, according to Artificial Analysis. The model also improved its rankings on text-to-video and image-to-video benchmarks.
X：Artificial Analysis (@artificialanlys)First-partyAI score 52Artificial Analysis published a leaderboard ranking search APIs, along with its methodology and the specific settings used for Keenable Search. All Search API results were produced with Stirrup, the team's open-source agent harness, so the evaluation can be reproduced.
X：Artificial Analysis (@artificialanlys)First-partyAI score 50
Other stories people are talking about
- Heat index 178NewAfter Anthropic Tightens Claude Access, Users Test Gemini 4 as Alternative2 sources
- Heat index 508RisingGoogle launches Gemini Decisions API as Perplexity open-sources Decider v1.313 sources
- Heat index 234RisingGemini 4 Carbon open to everyone, with 2x pricing from December4 sources
- Heat index 85NewGemini 网页版新对话页面卡顿的解决方案1 sources
- Heat index 449Google and Samsung open pre-orders for Gemini smart glasses7 sources
How is heat calculated?About the methodHide
Heat counts how many independent sources covered a story in the last 48 hours: one source counts once no matter how many posts it published, decaying with a 24-hour half-life. What ranks first is what many people are talking about.
This page aggregates public feeds. Headlines and summaries are machine-organized and remain the property of the original authors; verify important facts at the source. If you believe a headline or summary infringes your rights, email the contact address in the footer with the page URL and basis of your claim, and we will remove or replace it after review.
- Surge
- Discussion rising fast
- New
- First report within 6 hours
- Rising
- Still gathering discussion