Free
Free
Linum v2 2B-parameter weights are offered for download under Apache 2.0; hosted SaaS pricing is not listed.
- Access
- Linum v2 2B-parameter weights are offered for download under Apache 2.0; hosted SaaS pricing is not listed.
Video · Vendor site 2026-10-04
Linum is a two-person research lab training generative video models from scratch. The site showcases text-to-video generations, publishes field notes on model training, and distributes Linum v2 model weights under Apache 2.0 for download.
Researchers and developers experimenting with open generative video weights rather than teams buying a managed video SaaS.
Site focuses on research releases and weight downloads without a commercial self-serve product or support plan.
Linum demonstrates generative video samples and documents the road to Linum v3 through research posts on data filtering, training speed, and infrastructure.
What it can do
Unknown is different from unavailable. Each fact carries its own evidence.
| Capability | Value | Evidence | Checked |
|---|---|---|---|
| Open weights | Homepage links to Linum v2 2B params released under Apache 2.0. | Facts sourced | 2026-10-04 |
| Research focus | Tagline states a two-person lab training generative video models from scratch with published field notes. | Facts sourced | 2026-10-04 |
Understand the total cost
Free
Free
Linum v2 2B-parameter weights are offered for download under Apache 2.0; hosted SaaS pricing is not listed.
Ray3 on ray3.co is marketed as a reasoning video model for text-to-video and image-to-video generation with HDR output and draft mode. The site sells credit-based subscriptions for Ray3, Ray3 Modify, Raner, and related Luma-powered video tools.
Explore toolD-ID is an AI video platform that brings still photos to life with photorealistic speech and facial expressions. It specializes in generative AI video agents (Agents), digital person avatars, and API integration for customer service, corporate communications, and interactive marketing.
Explore toolHeyVid is an all-in-one AI video and image generator aggregating models such as Seedance, Veo, Sora, Kling, Runway, Midjourney, and Flux. It supports text, image, and video-to-video flows plus voice, music, avatars, and lip sync tools.
Explore toolThe practical questions
Linum is a two-person research lab training generative video models from scratch. The site showcases text-to-video generations, publishes field notes on model training, and distributes Linum v2 model weights under Apache 2.0 for download.
Linum v2 2B-parameter weights are offered for download under Apache 2.0; hosted SaaS pricing is not listed.. This record lists ongoing free access; check the plan limits before starting.
No paid monthly price is listed; this record treats the product as free to start. See the plan cards for entitlements, billing commitments and seat minimums.
Not confirmed. API access and subscription access may have different terms; consult the linked sources.
Site focuses on research releases and weight downloads without a commercial self-serve product or support plan.
No retained pricing changes yet. A current price alone does not establish a historical trend.
Reviewed vendor source
Read original source ↗Facts apply to the named version and check date. Send a sourced correction if something changed.