Tencent adds three video models to WorkRally: Rally-Video, Rally-Video-edit-2.0 and UpScale-rally with 4K output
Tencent has upgraded its WorkRally AI content-creation platform with three video models: Rally-Video for generation, Rally-Video-edit-2.0 for selective editing, and UpScale-rally for 2K and 4K super-resolution. Tencent claims UpScale-rally cuts processing time by 80% versus the previous version. Pricing was not disclosed.
Tencent has added three video models to WorkRally, its AI content-creation platform: Rally-Video, Rally-Video-edit-2.0 and UpScale-rally. The upgrade also extends the platform from animated short dramas to live-action short dramas, advertising and other creative video, according to Science and Technology Daily (November 6) and Shanghai Securities News (November 7), as relayed by Pandaily. The reports describe the upgrade as recent and give no exact release date.
The three models
Rally-Video is a generation model for film and animation work. It accepts images, video and untextured 3D models as references.
Rally-Video-edit-2.0 is an editing model that changes only selected elements of an existing clip, such as a character or the background, and leaves other areas untouched. It keeps the original duration, resolution and quality.
UpScale-rally is a super-resolution tool focused on detail in people and scenes. It offers speed and resolution modes and supports 2K and 4K output. The WorkRally product team says processing time is 80% shorter than the previous version. That figure comes from Tencent and has not been independently tested.
According to the reports, creators can choose among the models inside existing workflows based on task and budget.
Platform changes
Beyond the models, the upgrade includes:
- The agent-based creation entry point moved to the home page.
- A strengthened agent and a dedicated knowledge base.
- AI script generation and script diagnostics.
- Smart editing.
- One-click asset replacement and a new copyright-registration feature for AI-created assets (reported by Science and Technology Daily).
A WorkRally product lead said Tencent Video has turned years of production experience into knowledge bases, skills and vertical models that creators can call on directly. Topics include shot sequencing, lighting and camera movement. The stated aim is to reduce trial and error and rework so creators can focus on story.
Professional edition
A professional edition for animation studios, film crews and production companies remains invitation-only. It covers previsualization, concept design, scene compositing, CG asset production and visual effects.
Not yet disclosed
Pricing for the new models was not disclosed in the reports. Context window, parameter counts, benchmark scores and API availability were also not given.
What this means
The three models cover separate stages of a video workflow: generation from references, selective editing of existing footage, and resolution upscaling. Rally-Video-edit-2.0 is built to change chosen elements while preserving the clip's original duration, resolution and quality. UpScale-rally adds 2K and 4K output with speed and resolution modes. Tencent's 80% processing-time claim for UpScale-rally is untested, so its practical value is unverified.
The scope has widened from animated short dramas to live-action short dramas and advertising. The copyright-registration feature for AI-created assets addresses ownership of generated material, though the reports give no details on how it works. The professional edition stays invitation-only, so the new capabilities described for everyday creators are separate from the studio-oriented tools for previsualization and visual effects.
Related Articles
Tencent expands Hy4 inference cluster tenfold after WorkBuddy usage surge
Tencent says it has expanded the inference cluster serving Hy4 tenfold after many WorkBuddy users hit their quotas on the model's first day. Hy4 scores 35 on the Artificial Analysis Intelligence Index, second among Chinese models, and ranks 13th on Agent Arena.
Claude Personal Finance skips model training by default; ChatGPT Finances follows your opt-out setting
Anthropic's new Personal Finance feature for Claude does not use hub conversations for model training by default. OpenAI's ChatGPT Finances, which debuted in May, applies the user's existing training settings, where "Improve the model for everyone" is on by default. Engadget compared the two.
Huawei releases standalone HarmonyOS PC assistant (v11.7.9.300) with official Work office agent
Huawei has put its AI assistant in the HarmonyOS PC AppGallery as a standalone app, version 11.7.9.300, and officially released its Work office agent. IT Home reported on November 6 that Work moved beyond the internal testing it entered in September. It uses a membership model with AI points, and subscription prices have not been disclosed.
Meta brings Active Memory and Active Focus to all Muse and Meta AI users
Meta has introduced specialized versions of its existing models, trained using Muse Wonder. The versions bring Active Memory and Active Focus to all Muse and Meta AI users, according to The New Stack.
Comments
Loading...