第一项始终是图钉的来源。总体可信度是各份资料对上方所用开始和结束时间支持程度的加权平均；资料每比最新的一份旧 180 天，权重减半。
Google's[2][3] blog post 'Build with Veo 3.1 Lite, our most cost-effective video generation model' is stamped 31 March 2026 and says 'Rolling out today, you can access the model via paid tier on the Gemini API and Google AI Studio'; The Decoder[6] reports it the same day. Vertex AI lists 2 April 2026 for its preview.
Model page: text and image in, video with audio out; no 4K or Extension; the preview 'was shut down on October 22, 2026' with migration to Gemini Omni Flash.[1]
Lists veo-3.1-lite-generate-001 as a preview released 2 April 2026, against GA releases of 17 November 2025 for Veo 3.1 and 3.1 Fast.[1]
Arena lists Veo 3.1 Lite at Elo 950 (March 2026), below Veo 3.1 at 963 and Fast at 960.[1]
Nine one-try test clips from Lite on fal, taking 37 to 102 seconds each, judged softer than Fast on fine detail.[1]
有遗漏或错误吗？用你自己的话说明：能佐证此图钉的链接、不同的开始或结束日期及理由，或缺失、有误的信息。AI 会对照此图钉的来源进行核实，搜索更好的来源，并添加任何支持你说法的页面。图钉自身的来源仍然最重要。AI 也会查看图片：显示的是别的东西或显示效果差的图片会被移到后面或替换。
| 1,024 tokens |
| Not supported | 4K output; video extension |
|---|
| Speed | Same as Veo 3.1 Fast (Google) |
|---|
Gemini API video docs as archived 31 Mar 2026. Prompt limit and unsupported features from ai.google.dev/gemini-api/docs/models/veo-3.1-lite-generate-preview. The preview model was shut down on 22 Oct 2026. · 来源
| Resolution | Veo 3.1 Lite | Veo 3.1 Fast | Veo 3.1 |
|---|---|---|---|
| 720p | $0.05 | $0.15 | $0.40 |
| 1080p | $0.08 | $0.15 | $0.40 |
| 4K | Not offered | $0.35 | $0.60 |
From Google's launch pricing table as reported by The Decoder. Veo 3.1 Fast drops to $0.10 (720p), $0.12 (1080p) and $0.30 (4K) from 7 Apr 2026. Google gives Lite as less than 50% of Fast's cost at the same speed. · 来源
Veo 3.1 Lite keeps most of the family's look at $0.30 a clip and loses it on the details that matter for finished work.
Veo 3.1 Lite 950 ±17 95% interval 933 to 967 3,443 Mar 2026 $4.80 /min
The company doesn't specify quality differences between the three tiers.
