LOADING DATASHEET
LOADING DATASHEET
by Lightricks
RANKED #16 OF 22 VIDEO · OVERALL #152 OF 6,543 · VIBE SCORE 6.0 · 2,737 VOICES
A fast, open-weights AI video generator designed for creative projects.
119 mentions
79 mentions
26 weeks · 2,955 voices
HONEYMOON FADING
Early praise is cooling off in recent weeks.
Open weights. You can use a hosted service, or download it and run it yourself, free.
Open weights: download LTX 2.5 once and it's yours. No subscription, no rate limits, works offline.
22B parameters, quantized FP8 / INT8 transformer checkpoints are around 12 to 24 GB
NVIDIA GeForce RTX 4060 Ti 16GB
16 GB VRAM
~$450INT8 distilled checkpoint with 4-bit Gemma text encoder and CPU offload, 720p at lower frame counts
NVIDIA GeForce RTX 4090 24GB
24 GB VRAM
~$1750FP8 distilled or dev transformer with standard text encoder, fast 720p to 1080p generation
NVIDIA RTX 6000 Ada 48GB
48 GB VRAM
~$6800Full BF16 precision model, high resolution, multishot generation without memory offloading
The three cards above are researched picks. Search the machine you actually have — we only say yes if it should reply at a usable speed, not just load the weights and crawl.
HARDWARE PICKS & PRICES ARE RESEARCHED FROM THE LIVE WEB AND REFRESHED AUTOMATICALLY. TREAT THEM AS BALLPARK, NOT GOSPEL.
Everyday jobs on each of those rigs: how long LTX 2.5 takes, and how good the crowd says it is at that kind of work.
| THE JOB | BARE MINIMUMNVIDIA GeForce RTX 4060 Ti 16GB | SWEET SPOTNVIDIA GeForce RTX 4090 24GB | FULL POWERNVIDIA RTX 6000 Ada 48GB |
|---|---|---|---|
| Make a video clipfive seconds of footage | —★★★★★ | —★★★★★ | —★★★★★ |
STARS ARE THE COMMUNITY SCORE FOR THAT KIND OF WORK, DOCKED FOR HOW SQUASHED THE WEIGHTS GET AT EACH TIER. A DASH MEANS NO SPEED WAS RESEARCHED FOR THAT RIG. TIMES ARE COMPUTED FROM RESEARCHED SPEEDS. BALLPARK, NOT GOSPEL.
Users love LTX 2.5 for its fast generation, great visual results, and good value. However, some note that it can occasionally ignore parts of your prompt.
448 POSTS · 2289 COMMENTS · REDDIT · X · GITHUB · BLUESKY · DEV.TO · DEV FORUMS · BLOG
6 thumbs up · 0 thumbs down
0 thumbs up · 5 thumbs down
3 thumbs up · 1 thumbs down
3 thumbs up · 1 thumbs down
1 thumbs up · 2 thumbs down
2 thumbs up · 1 thumbs down
Checked picks first: tags say why you would switch, stars say how fully each one stands in for LTX 2.5.
LTX-2.5 is Here
LTX-2 Updates
GLM-5.3 shows how much capability may still be hiding inside today’s largest base models and how relevant post-training
LTX 2.5. on 3060/16gb ram, 0.5mp 10 second video took 180 seconds to generate.
LTX 2.5 comparison table vs Minimax H3 is a pathetic bullshit
LTX 2.5 just dropped today
ReDetail: Upscale MiniMax H3 renders with the LTX-2.5 video upscaler on 24GB+ VRAM
LTX-2.5 is out!
LTX 2.5 x2 upscaler for Minimax H3 on 4090
LTX 2.5 is really Fast 🔥
Made this with LTX-2.5 (i2v)
LTX 2.5 Is Disappointing