Lightricks: LTX-2.5 2x
lightricks/ltx-2.5-2xHugging FaceVideoaudio-to-audioaudio-to-videodiffusion-single-fileimage-text-to-audio-videoimage-text-to-videoimage-to-audio-videoimage-to-videotext-to-audiotext-to-audio-videotext-to-videovideo-to-audiovideo-to-video
Reference inputs & limits
Accepted uploads and guidance from each provider. Limits apply to one generation.
ImpCC · video generation · up to 3 files · 20 MiB total
image as first_frame (up to 1), last_frame (up to 1) · image/png, image/jpeg, image/webp
audio as reference (up to 1) · audio/wav, audio/mpeg, audio/flac, audio/ogg, audio/mp4, audio/webm
video as canny (up to 1), depth (up to 1), openpose (up to 1) · video/mp4, video/webm
Control passes served together: depth; canny; openpose
Combine at most three items in one generation: one or two images, the frame the clip starts on and the frame it ends on; one control video; and one audio file. The video is an already rendered depth, edge or pose pass whose structure the clip keeps, one pass per generation, read at an attention of 0.7 and never noised. Audio supplies the soundtrack and requires audio output enabled; sound from the video file is not used. Audio and video inputs must be at most 60 seconds; images and video frames must contain at most 16,777,216 pixels. Visual references are scaled to cover the output frame and centre-cropped to it, so a pass at another aspect loses its edges rather than gaining bars that read as far depth. Clips are trimmed or extended to the selected output length: video repeats its last frame and audio pads with silence. The clip is rendered in two stages, composed at half the output frame and refined at the full one, which reaches twice the frame lightricks/ltx-2.5 serves; every reference conditions the composing stage.
Versions
Other models in this family. Each has its own providers and prices.
- Lightricks: LTX-2.5lightricks/ltx-2.5baseFree in— out1K2 endpoints
Providers
Live endpoints for this model. Response is the time from request to result; uptime is measured over the last day and week. Nothing here streams, so there is no first token to time and no token rate to report.
| Provider | Prompt limit | Sizes | Clips | Per video second | Response | Capacity | Uptime 1d | Uptime 7d | Health | Data |
|---|---|---|---|---|---|---|---|---|---|---|
| ImpCCint890% offCA | 1K | ≤2.35 MP · 512–2432px · ×128 | ≤9s · 24, 25 fps · 8n+1 · audio | $0.0030 | 14.5s30m | 1/1 free | 100.0% | 100.0% | Healthy | ZDR |
Pricing
USD per output video second.
| Provider | Per video second |
|---|---|
| ImpCCint890% off | $0.0030 |
Price history
Price · USD per output video second
Performance
Percentiles cover the labelled window: the last 30 minutes when it has enough samples, otherwise the last day. Response is the time in seconds from request to result. This model answers whole rather than streaming, so there is no first token to time and no rate between tokens to report — response time is the whole measurement there is.
| Provider | Response p50 | p75 | p90 | p99 |
|---|---|---|---|---|
| ImpCCint8 | 14.5s30m | 18.4s | 20.0s | 26.3s |
Uptime
Share of successful requests and probes over the last day and week.
- ImpCCint81d 100.0%7d 100.0%Healthy
Availability by hour · last 7 days
Share of successful requests and probes per hour and provider. The line spans hours with nothing to measure; where samples are sparse, a dot marks each hour actually measured.
Activity
Megapixels per UTC day over the last 30 days. Every billed request counts, the platform's own paid probe traffic included.
job=$(curl -s https://api-staging.impossiblecarrot.cc/api/v1/videos \ -H "Authorization: Bearer $HEIMDALL_API_KEY" \ -H "Content-Type: application/json" \ -d '{ "model": "lightricks/ltx-2.5-2x", "prompt": "a paper boat drifting across a rain puddle, neon reflections", "size": "512x512", "duration": 1 }')job_id=$(echo "$job" | jq -r '.id')# Poll until the job reaches a terminal status.while true; do status=$(curl -s https://api-staging.impossiblecarrot.cc/api/v1/videos/$job_id -H "Authorization: Bearer $HEIMDALL_API_KEY") state=$(echo "$status" | jq -r '.status') [ "$state" = "pending" ] || [ "$state" = "in_progress" ] || break sleep 2donecurl -s "https://api-staging.impossiblecarrot.cc/api/v1/videos/$job_id/content?index=0" \ -H "Authorization: Bearer $HEIMDALL_API_KEY" \ -o out.mp4Set HEIMDALL_API_KEY to a key from your account. See the quickstart for routing suffixes and provider preferences.
About
LTX-2.5 rendered in two stages: the clip is sampled at half the requested frame, the latent is upsampled 2x by the LTX-2.5 latent spatial upscaler, and a short refine pass finishes it at the requested frame, up to 2.36 megapixels on a 128 grid with no edge over 2432 pixels. Clips come to a multiple of eight frames plus one, so at 25 fps the whole-second lengths are 1 and 9; an omitted length renders 3.88 seconds at 2048x1152.
- Author
- Lightricks
- Family
- lightricks/ltx-2.5
- Weights
- Lightricks/LTX-2.5
- Parameters
- crf (0–51), duration (1, 9), first_frame_compression (0–100), first_frame_strength (0–1), fps (24, 25), generate_audio (true/false), guidance_scale (0–20), input_references (list), reference_strength (0.01–1), refine_guidance_scale (0–20), refine_sampler (45 choices), refine_sigmas (list), sampler (45 choices), seconds (any value), seed (0–2147483647), sigmas (list), size (any value)
- Size
- —
- Tokenizer
- —
- Available until
- —
From the model card
What the repository states about these weights, read from Lightricks/LTX-2.5 on Oct 11, 2026.
- Languages
- English, German, Spanish, French, Japanese, Korean, Chinese, Italian, and 1 more
- Papers
- 2601.03233
- Datasets
- —
- Revision
- 49b093aOct 11, 2026
- Downloads
- 1.7M
- Likes
- 7.3K
- Availability
- Gated — the repository serves its files only once its licence is accepted.