Lightricks Open-Sources LTX-2.5: Generates 10s Video in 6.8s
xiaohu · x · 2026-08-13
Lightricks has released and open-sourced LTX-2.5, a 22B parameter video generation model that can produce a 10-second video clip in just 6.8 seconds, significantly outpacing Veo 3.1.
Key Features:
- Unified Audio-Visual Generation: Video and audio are generated natively within the same model without external tools.
- Native Multi-Shot: Generates multi-shot sequences in one pass, maintaining consistency in characters, scenes, and sound across shots.
- Local Optimization: Optimized VRAM usage for RTX GPUs, includes NVFP4 and int8 quantized versions.
Weights are available on Hugging Face with day-one ComfyUI support. The open-source license is fee-free for companies with under $10M in annual revenue.
Related event: Lightricks Open-Sources LTX-2.5 Video Model with Native Multishot(24 posts)→
More from Infra
- Zero's next-gen IVM engine runs in a single DO, live-querying a 100GiB R2 database — threepointone · 2026-10-16
- Meme: driving to the chip factory in 2028 captures AI's compute anxiety — beffjezos · 2026-10-16
- One volunteer maintains the time zone file behind 4 billion phones — danbri · 2026-10-16
- DeepSeek V4.1 Flash on 4x DGX Spark: weights alone are 511GB, the 494 tok/s claim doesn't fit — Aroochacha · 2026-10-16
- Open-sourced Intel GPU Vulkan inference upgrades: 2x ingestion, 60%+ faster inference — Fun-Aardvark-1143 · 2026-10-16
- ELSFP pain points: >3dB insertion loss, costly PMF, harsh packaging requirements — jwt0625 · 2026-10-16