Veo 3.1
Google's Veo 3.1 cinematic video model with synchronized audio, released in public preview in October 2025 and offered in standard, Fast and Lite tiers. It is now well behind the 2026 video models on both LMArena and Artificial Analysis.
GoogleVideo modelReleased October 2025Data checked October 18, 2026
Where people rank it
#9
Creating video (tied), 13,741 votes
From blind votes on LMArena, where people pick the better of two anonymous answers. See the full tables on the AI leaderboard.
Strengths and weaknesses
- Large LMArena samples, so its scores are stable
- Available in the Gemini API in standard, Fast and Lite tiers
- Every tier generates video with audio
- Expensive at standard quality ($0.40 per second)
- Lowest LMArena text-to-video score among the video models here
Best for
- High-fidelity cinematic shots
- Videos with synchronized sound
Every score we have
Each bar compares this model with the best result any model on this page has on the same test.
- LMArena text to video1,364.2Arena score 1364.2 (95% CI 1349.8 to 1378.6); 13,741 votes; entry "veo-3.1-audio"; position 12 of 48 on the full board
- LMArena image to video1,398.1Arena score 1398.1 (95% CI 1387.6 to 1408.6); 25,116 votes; entry "veo-3.1-audio"; position 13 of 48 on the full board
- Artificial Analysis video arena967.5Elo 967.5 (95% CI 949.5 to 985.5); 3,043 appearances; position 19 of 32 on the Artificial Analysis text-to-video arena
- Price per minute of video (estimate)$24$24.00 per minute of video (Artificial Analysis estimate)
Compare with
- Gemini Omni FlashGoogle's Omni video model, first shown at I/O 2026, which generates and edits video with native audio; the 1.1 Flash version became generally available in the Gemini API on August 27, 2026.
- Wan 3.0Alibaba's Wan 3.0 video model, offered through Alibaba Cloud Model Studio, with clips of up to 30 seconds at up to 1080p.
- MiniMax H3MiniMax's open-weight omni-modal video model, which generates 4 to 15 second clips with native stereo audio from text, images, video and audio.
- FLUX 3 VideoBlack Forest Labs' FLUX 3 video model, which generates video with synchronized audio from text, keyframes or an existing clip.