Gemini Omni 1.2 Flash
369 points
• 6 days ago
• Article
Link
Google 已推出 Gemini Omni 1.2 Flash，这是其视频生成模型的重大更新，目前已通过 Gemini API 和 Google Cloud 全面开放使用。此次发布带来了可生成最长 15 秒视频内容的能力，让创作者和企业有更充足的空间去打造更完整的叙事、产品讲解以及广告短片。该模型在参考一致性方面也实现了显著提升，能够确保角色、产品和风格在不同摄像机角度下保持一致，这对于虚拟试穿体验和品牌化内容等专业级任务至关重要。
这一新迭代的核心优势在于强调更具成本效益的高质量制作。 Google 将价格相较上一版本下调了 30%：目前 720p 输出的定价为每秒 0.07 美元。工作流程面向规模化需求进行了优化，用户可以先用较低分辨率起草内容，再针对特定细节进行迭代修改，而无需反复重生成整段视频。新的多轮对话式编辑能力进一步提升了效率：用户只需通过连续、逐步的提示，就能对背景、光照或产品摆放等元素进行精细调整。
在基础生成之外，Gemini Omni 1.2 Flash 还对动作与语音进行了技术强化。模型生成的对话听起来更自然，唇形同步更清晰，并支持区分的多说话人条件设定。它还能提供对电影级摄像机运动的更强控制，使得如 FPV 飞行穿越与镜头拉远这类连续镜头更顺畅，同时减少跳切并提升画面稳定性。这些在一致性与可控性上的改进，有助于从最初的创意想法更顺利地落到精修后的成品结果。
系统内置了安全与可追责机制，所有生成内容都将带有肉眼不可察的 SynthID 水印。该模型已被多类机构采用，包括创意代理机构、 VFX 从业者以及电商平台，用于简化并加速视频生产流程。通过将企业级基础设施与这些新的生成能力结合起来，Google 旨在为专业团队提供可靠工具，帮助他们更高效地制作、编辑并扩展视频内容。
Google has launched Gemini Omni 1.2 Flash, a major update to its video generation model that is now generally available through the Gemini API and Google Cloud. This release introduces the ability to generate video content up to 15 seconds long, providing creators and businesses with more room to build comprehensive narratives, product explainers, and advertising spots. The model also features significant improvements in reference adherence, ensuring that characters, products, and styles remain consistent across camera angles, which is critical for professional-grade tasks like virtual try-on experiences and branded content.
A primary advantage of this new iteration is the emphasis on cost-effective, high-quality production. Google has reduced the pricing by 30% compared to the previous version, now set at $0.07 per second for 720p output. The workflow is designed for scalability, allowing users to draft content at lower resolutions and iterate through specific details rather than re-generating entire clips. This efficiency is further bolstered by new multi-turn conversational editing capabilities, which let users refine aspects like backgrounds, lighting, or product placement through simple, successive prompts.
Beyond basic generation, Gemini Omni 1.2 Flash brings technical refinements to motion and speech. The model produces more natural-sounding dialogue with clearer lip-syncing and supports distinct multi-speaker conditioning. It also offers greater control over cinematic camera movements, enabling continuous shots like FPV flythroughs and zoom-outs with fewer jump cuts and improved visual stability. These advancements in consistency and control help bridge the gap between initial creative concepts and polished final outputs.
Safety and accountability are integrated into the system, with all generated content featuring an imperceptible SynthID watermark. The model is already being utilized by diverse organizations, ranging from creative agencies and VFX artists to e-commerce platforms, to streamline their production pipelines. By combining enterprise-grade infrastructure with these new generative capabilities, Google aims to provide a reliable tool for professional teams to produce, edit, and scale video content efficiently.
256 comments • Comments Link
• Google 的 AI 产品组合陷入混乱且令人困惑的生态：产品之间重叠、命名口径不一致（例如 AI Studio 、 Vertex 以及各种 Gemini 层级），让用户难以判断自己到底为哪些能力付费、以及如何访问特定模型。
• 客户摩擦问题十分严重。有报告称支付流程出现中断、身份验证反复卡住进入无休止循环，以及组织账户政策过于严格，导致 Workspace 下的个人用户无法使用面向消费者的 AI 计划。
• Google 开发者界面的用户体验遭到广泛吐槽，被认为比 Anthropic 或 OpenAI 等竞争对手更不直观、更不可靠。结果是，许多开发者不得不依赖 OpenRouter 等第三方聚合器来绕开 Google 的直接基础设施。
• 内部官僚主义被形容为"根深蒂固"，且容易陷入由委员会主导的停滞局面。批评者认为，拥有大量资源的公司所产出的软件，往往比更小、更灵活的竞争对手更像是没打磨好或功能不完整，主要原因就在于这种组织方式。
• 对尖端模型的访问发布节奏并不一致，导致付费的"Pro"订阅用户被困在较旧的模型版本上，或频繁遭遇难以解释的锁定与降级，被迫退回到更低层级的模型。
• 公司内部被认为存在深层次的隔离感，批评人士称 Google 的组织结构阻碍了他们理解并解决开发者与高阶用户的真实需求。
• "Omni"和"Flash"的品牌策略被认为让人困惑而非具有描述性。用户对缺乏清晰、统一的命名规范来区分速度、能力与模型用途感到不满。
• 尽管有些用户认为 Google 用于编码的"agentic"工具有价值，但这种成功常常被质量不佳的"harness"所掩盖：即便集成工具本身也会出问题、引入 Bug，并且很难在专业工作流程中保持可靠性。
• 该公司被指更重视"华丽、演示友好"的功能，而非把核心实用工具打磨到位，尤其是在办公套件的集成方面。相比之下，竞争对手的工具通常因性能与可靠性更出色而更受青睐。
• 关于账户被封禁的持续担忧，源自 Google 严格且有时过度敏感的自动化安全政策，这使用户难以把 Gemini 模型集成到敏感或关键的个人工作流程中。
此次讨论反映出一种普遍共识：尽管 Google 拥有世界级工程人才和重要的 AI 研究，但在产品包装与交付执行上存在明显短板。用户对该公司无法为其技术提供稳定、易访问且透明的界面表达强烈不满。也因此，许多人认为该组织更关注内部碎片化与"掩体式"开发，而不是服务外部客户。最终，Google 的技术能力与其糟糕的 UX/ 入门体验之间的落差，促使不少高阶用户转向竞争对手或第三方网关，并进一步强化了一种舆论：Google 未能把研究优势转化为真正可用、可落地的软件产品。 • Google's AI offerings are plagued by a chaotic and confusing ecosystem where overlapping, inconsistently named products—such as AI Studio, Vertex, and various Gemini tiers—make it difficult for users to understand what they are paying for or how to access specific models.
• Customer friction is extreme, with reports of broken payment flows, endless identity verification loops, and restrictive organizational account policies that make it impossible for individual users on Workspace accounts to access consumer-grade AI plans.
• The user experience within Google's developer interfaces is widely criticized as being significantly less intuitive and reliable than competitors like Anthropic or OpenAI, leading many developers to rely on third-party aggregators like OpenRouter to bypass the direct Google infrastructure.
• Internal bureaucratic challenges, described as "entrenched" and prone to committee-led stagnation, are perceived as the primary reason why a company with vast resources produces software that often feels unpolished or functionally deficient compared to smaller, more agile competitors.
• Access to cutting-edge models is often inconsistently rolled out, leaving paying "Pro" subscribers stuck on outdated model versions or facing frequent, unexplained lockouts and downgrades to lower-tier models.
• There is a profound sense of isolation within the company, with critics suggesting that Google's organizational structure prevents them from understanding and addressing the real-world needs of developers and power users.
• The "Omni" and "Flash" branding strategies are viewed as confusing rather than descriptive, with users expressing frustration over the lack of a clear, standardized naming convention that would help distinguish between speed, capability, and model purpose.
• While some users find value in Google's "agentic" tools for coding, these successes are often overshadowed by poor "harness" quality, where the integration tools themselves fail, introduce bugs, or struggle with reliability in professional workflows.
• The company is seen as prioritizing "flashy" demo-ready features over the refinement of core utility, particularly in office-suite integrations where competitors' tools are often preferred for their superior performance and reliability.
• Persistent reports of account-ban fears—stemming from Google's strict and sometimes over-sensitive automated safety policies—discourage users from integrating Gemini models into sensitive or critical personal workflows.
The discussion reflects a broad consensus that while Google possesses world-class engineering talent and significant AI research, its execution in product packaging and delivery is deeply flawed. Users express deep frustration with the company's inability to provide a stable, accessible, and transparent interface for its technology, leading many to feel that the organization is more interested in internal fragmentation and "bunker" development than in serving external customers. Ultimately, the perceived gap between Google's technical capability and its poor UX/onboarding experience has driven many power users toward competitors or third-party gateways, fostering a sentiment that the company is failing to translate its research dominance into practical, usable software products.