# LTX 2.3 Trainer (V2) - Backward Video Extension API

> Usposobi LoRA, ki video podaljša nazaj v času in ustvari uvod pred končnim posnetkom

Usposobi LoRA za podaljšanje videa nazaj, ki ustvari uvod pred podanim končnim posnetkom. `conditioning_frames` določa število končnih sličic, uporabljenih kot čisti zaključek (8 do 121, večkratnik 8). Ko je `with_audio` omogočen, se zvok podaljša skupaj z videom, zato mora vsak učni in validacijski posnetek vsebovati zvok. Podatki se posredujejo kot zip prek `training_data_url`.

- Model id: `ltx23-trainer-v2/extend-suffix`
- Made by: Lightricks
- Category: Training
- Output: file
- Price: $0.0061 / step (cheapest run $12.20 (12200 API credits))
- Run: `POST https://api.doitong.com/v1/run/ltx23-trainer-v2/extend-suffix`
- Quote: `POST https://api.doitong.com/v1/quote/ltx23-trainer-v2/extend-suffix`
- Docs page: https://doitong.com/sl/developers/models/ltx23-trainer-v2/extend-suffix

## Pricing

Price: **$0.0061 / step**. Billed per training step (`number_of_steps`).

The full example input below costs $12.20 = 12200 API credits (2000 steps × 6.1 API credits).

Prices are the upstream price in USD, paid from the API balance in API credits (1000 API credits = 1 USD). `POST https://api.doitong.com/v1/quote/ltx23-trainer-v2/extend-suffix` with the same JSON body returns the exact price for that input; the quote is exactly what a run charges. Invalid input is never charged and failed or canceled requests are refunded automatically.

## Input parameters

| Parameter | Canonical name | Type | Required | Default | Allowed values / range | Description |
|---|---|---|---|---|---|---|
| `training_data_url` |  | string | yes |  |  | URL datoteke `.zip` s podatki za učenje, organiziranimi glede na izbrani način učenja. Vsaka predstavnostna datoteka ima lahko spremljajoč opis `.txt` z enakim osnovnim imenom. |
| `rank` |  | enum (integer) | no | `32` | `8`, `16`, `32`, `64`, `128` | Rang prilagoditve LoRA, pri katerem večja nastavitev pomeni večjo porabo pomnilnika in večjo zmogljivost. |
| `number_of_steps` |  | integer | no | `2000` | 100 – 10,000 | Skupno število korakov učenja. |
| `learning_rate` |  | number | no | `0.0002` | 0.000001 – 1 | Hitrost učenja pri optimizaciji. |
| `number_of_frames` |  | integer | no | `89` | 9 – 121 | Število sličic za vsak učni vzorec. Deljenje števila z 8 mora pustiti ostanek 1; primeri so 1, 9, 17, 25, 33, 41, 49, 57, 65, 73, 81, 89, 97. |
| `frame_rate` |  | integer | no | `24` | 8 – 60 | Ciljna hitrost sličic učnega videa v sličicah na sekundo. Naravna hitrost za LTX-2.3 je 24. |
| `resolution` |  | enum (string) | no | `medium` | `low`, `medium`, `high` | Ločljivost učenja. Povečanje ločljivosti poveča zahteve po pomnilniku. |
| `aspect_ratio` |  | enum (string) | no | `1:1` | `16:9`, `1:1`, `9:16` | Razmerje stranic pri učenju. |
| `trigger_phrase` |  | string | no | `` |  | Besedna zveza za aktivacijo sloga LoRA, vstavljena na začetek opisov med učenjem. |
| `auto_scale_input` |  | boolean | no | `false` |  | Samodejna prilagoditev videa zahtevanemu FPS in številu sličic, ko je omogočena; nabori slik ostanejo nespremenjeni. |
| `split_input_into_scenes` |  | boolean | no | `true` |  | Razdeljevanje prizorov pri videih, ki presegajo prag trajanja, omogočeno, ko je true. |
| `split_input_duration_threshold` |  | number | no | `30` | 1 – 60 | Sprožilec razdeljevanja prizorov, izražen kot trajanje v sekundah: razdelijo se samo videi, daljši od praga. |
| `debug_dataset` |  | boolean | no | `false` |  | Način pregleda predobdelave za preverjanje videov, slik in opisov pred celotnim učenjem. Z omogočanjem trener vrne obdelani učni nabor v arhivu, ki ga je mogoče prenesti in ročno pregledati. |
| `with_audio` |  | boolean | no | `false` |  | Način razširitve: false, privzeto, razširi samo video in vrne izhod brez zvoka. Pri true se zvok in video nadaljujeta iz skupnega končnega okna, vsak učni in validacijski izrezek pa mora imeti zvočni posnetek. |
| `audio_normalize` |  | boolean | no | `true` |  | Normalizacija vrhov zvoka za ohranjanje enakomernih ravni v celotnem naboru. |
| `audio_preserve_pitch` |  | boolean | no | `true` |  | Ohranjanje višine tona pri prilagajanju zvoka dolžini videa namesto njegovega krajšanja ali dodajanja tišine. |
| `conditioning_frames` |  | integer | no | `8` | 8 – 121 | Število sličic čiste končnice, preslikano v latentni `temporal_boundary` kot `conditioning_frames//8`. Za generiranje uvoda mora imeti izrezek več latentnih sličic kot ta meja. Končnica uporablja končne slikovne sličice, z `with_audio` pogojuje tudi zvok in zahteva število, deljivo z 8, na primer 8, 16 ali 24. |
| `validation` |  | array of object {prompt, video_url} | no | `[]` | ≤ 2 items | Validacijski podatki za razširitev videa nazaj, posredovani kot seznam vnosov s pozivom in videom za razširitev. |
| `validation[].prompt` |  | string | yes |  |  | Poziv, uporabljen med preverjanjem. |
| `validation[].video_url` |  | string (video URL) | yes |  | ≥ 1 characters | URL videa za razširitev vzdolž časovne osi. |
| `validation_negative_prompt` |  | string | no | `blurry, out of focus, overexposed, underexposed, low contrast, washed out colors, excessive noise, grainy texture, poor lighting, flickering, motion blur, distorted proportions, unnatural skin tones, deformed facial features, asymmetrical face, missing facial features, extra limbs, disfigured hands, wrong hand count, artifacts around text, inconsistent perspective, camera shake, incorrect depth of field, background too sharp, background clutter, distracting reflections, harsh shadows, inconsistent lighting direction, color banding, cartoonish rendering, 3D CGI look, unrealistic materials, uncanny valley effect, incorrect ethnicity, wrong gender, exaggerated expressions, wrong gaze direction, mismatched lip sync, silent or muted audio, distorted voice, robotic voice, echo, background noise, off-sync audio, incorrect dialogue, added dialogue, repetitive speech, jittery movement, awkward pauses, incorrect timing, unnatural transitions, inconsistent framing, tilted camera, flat lighting, inconsistent tone, cinematic oversaturation, stylized filters, or AI artifacts.` |  | Negativni poziv, uporabljen med preverjanjem, ki uporablja predogledni proces z eno stopnjo. Ker ta proces približa destilirano produkcijsko sklepanje z dvema stopnjama, se njegovo vodenje in kakovost predogleda razlikujeta od končnih rezultatov. |
| `validation_number_of_frames` |  | integer | no | `89` | 9 – 121 | Število sličic za vsak video preverjanja. |
| `validation_frame_rate` |  | integer | no | `24` | 8 – 60 | Ciljna hitrost sličic za videe preverjanja v sličicah na sekundo; LTX-2.3 izvorno deluje pri 24. |
| `validation_resolution` |  | enum (string) | no | `high` | `low`, `medium`, `high` | Ločljivost, uporabljena med preverjanjem. |
| `validation_aspect_ratio` |  | enum (string) | no | `1:1` | `16:9`, `1:1`, `9:16` | Razmerje stranic, uporabljeno med preverjanjem. |
| `stg_scale` |  | number | no | `1` | 0 – 3 | Merilo prostorsko-časovnega vodenja (STG), pri čemer je priporočena vrednost 1.0, vrednost 0.0 pa STG izklopi. |

## Example input

Required parameters only:

```json
{
  "training_data_url": "https://example.com/your-file.zip"
}
```

A fuller example:

```json
{
  "training_data_url": "https://example.com/your-file.zip"
}
```

## Run it

Set `DOITONG_API_KEY` to a key from https://doitong.com/developers/dashboard#keys. `?wait=20` returns the finished request in the same call when it is ready within 20 seconds; otherwise poll `GET /requests/{id}` (it also accepts `?wait=`), or pass `webhook_url`.

### curl

```bash
# Run the model; wait up to 20 s for the finished result in the same call
curl -X POST "https://api.doitong.com/v1/run/ltx23-trainer-v2/extend-suffix?wait=20" \
  -H "Authorization: Bearer $DOITONG_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "training_data_url": "https://example.com/your-file.zip"
  }'

# Still "queued" or "running"? Poll with the id from the response:
curl "https://api.doitong.com/v1/requests/REQUEST_ID?wait=20" \
  -H "Authorization: Bearer $DOITONG_API_KEY"
```

### JavaScript

```javascript
const API = "https://api.doitong.com/v1";
const headers = {
  "Authorization": "Bearer " + process.env.DOITONG_API_KEY,
  "Content-Type": "application/json",
};

// Run the model; wait up to 20 s for the finished result in the same call
let res = await fetch(API + "/run/ltx23-trainer-v2/extend-suffix?wait=20", {
  method: "POST",
  headers,
  body: JSON.stringify({
    "training_data_url": "https://example.com/your-file.zip"
  }),
});
let request = await res.json();
if (!res.ok) throw new Error(request.error.code + ": " + request.error.message);

// Not finished yet: long-poll until it is
while (request.status === "queued" || request.status === "running") {
  res = await fetch(API + "/requests/" + request.id + "?wait=20", { headers });
  request = await res.json();
}
if (request.status !== "succeeded") throw new Error(JSON.stringify(request.error));
console.log(request.outputs.map((o) => o.url), request.cost.credits + " API credits ($" + request.cost.usd + ")");
```

### Python

```python
import os
import requests

API = "https://api.doitong.com/v1"
headers = {"Authorization": "Bearer " + os.environ["DOITONG_API_KEY"]}

payload = {
    "training_data_url": "https://example.com/your-file.zip",
}

# Run the model; wait up to 20 s for the finished result in the same call
r = requests.post(API + "/run/ltx23-trainer-v2/extend-suffix", params={"wait": 20}, json=payload, headers=headers)
request = r.json()
if not r.ok:
    raise RuntimeError(request["error"]["code"] + ": " + request["error"]["message"])

# Not finished yet: long-poll until it is
while request["status"] in ("queued", "running"):
    request = requests.get(API + "/requests/" + request["id"], params={"wait": 20}, headers=headers).json()
if request["status"] != "succeeded":
    raise RuntimeError(request["error"])
print([o["url"] for o in request["outputs"]], request["cost"]["credits"], "API credits", "$" + request["cost"]["usd"])
```

### MCP

The MCP server at `https://api.doitong.com/v1/mcp` exposes the tools `search_models`, `get_model`, `quote`, `run`, `get_request`, `cancel_request`, `list_requests`, `create_upload` and `get_account`. A `run` call for this model:

```json
{
  "jsonrpc": "2.0",
  "id": 1,
  "method": "tools/call",
  "params": {
    "name": "run",
    "arguments": {
      "model_id": "ltx23-trainer-v2/extend-suffix",
      "input": {
        "training_data_url": "https://example.com/your-file.zip"
      },
      "wait_seconds": 20
    }
  }
}
```

## Response

Every call returns a request object. `status` is one of `queued`, `running`, `succeeded`, `failed`, `canceled`. `outputs` lists the generated files with URLs on our storage (kept forever, or 7 days with `?storage=temporary`; `storage` says which); `output` is the model's own result object with the same URLs.

```json
{
  "id": "req_8k2m4q7x1z9c3v5b6n0p2r4t",
  "object": "request",
  "model": "ltx23-trainer-v2/extend-suffix",
  "status": "succeeded",
  "created_at": "2026-10-20T12:00:00Z",
  "started_at": "2026-10-20T12:00:02Z",
  "completed_at": "2026-10-20T12:00:09Z",
  "queue_position": null,
  "cost": {
    "credits": 12200,
    "usd": "12.20",
    "charged": 12200,
    "refunded": 0
  },
  "input": {
    "training_data_url": "https://example.com/your-file.zip"
  },
  "output": {
    "video": {
      "url": "https://content.doitong.com/dev-api/2026/10/20/req_8k2m4q7x1z9c3v5b6n0p2r4t/0-output.jpeg",
      "content_type": "image/jpeg",
      "file_name": "0-output.jpeg",
      "file_size": 1843200
    },
    "audio": {
      "url": "https://content.doitong.com/dev-api/2026/10/20/req_8k2m4q7x1z9c3v5b6n0p2r4t/0-output.jpeg",
      "content_type": "image/jpeg",
      "file_name": "0-output.jpeg",
      "file_size": 1843200
    },
    "lora_file": {
      "url": "https://content.doitong.com/dev-api/2026/10/20/req_8k2m4q7x1z9c3v5b6n0p2r4t/0-output.jpeg",
      "content_type": "image/jpeg",
      "file_name": "0-output.jpeg",
      "file_size": 1843200
    },
    "config_file": {
      "url": "https://content.doitong.com/dev-api/2026/10/20/req_8k2m4q7x1z9c3v5b6n0p2r4t/0-output.jpeg",
      "content_type": "image/jpeg",
      "file_name": "0-output.jpeg",
      "file_size": 1843200
    },
    "debug_dataset": {
      "url": "https://content.doitong.com/dev-api/2026/10/20/req_8k2m4q7x1z9c3v5b6n0p2r4t/0-output.jpeg",
      "content_type": "image/jpeg",
      "file_name": "0-output.jpeg",
      "file_size": 1843200
    }
  },
  "outputs": [
    {
      "type": "file",
      "url": "https://content.doitong.com/dev-api/2026/10/20/req_8k2m4q7x1z9c3v5b6n0p2r4t/0-output.jpeg",
      "content_type": "image/jpeg",
      "file_size": 1843200,
      "stored": true
    }
  ],
  "error": null,
  "metrics": {
    "inference_seconds": 6.8
  },
  "urls": {
    "self": "https://api.doitong.com/v1/requests/req_8k2m4q7x1z9c3v5b6n0p2r4t",
    "cancel": "https://api.doitong.com/v1/requests/req_8k2m4q7x1z9c3v5b6n0p2r4t/cancel"
  }
}
```

### Output fields (`output`)

| Field | Type | Description |
|---|---|---|
| `video` | object {url, content_type, file_name, file_size} \| null | Združen video predogled preverjanja, kadar je na voljo za končne točke, ki ustvarjajo video. |
| `audio` | object {url, content_type, file_name, file_size} \| null | Združen zvok preverjanja, na voljo kot predogled za te poti samo z zvokom: /audio-inpaint, /audio-extend-suffix, /audio-extend-prefix, /t2a, /a2a in /v2a. |
| `lora_file` | object {url, content_type, file_name, file_size} | URL uteži naučenega LoRA v obliki .safetensors. |
| `config_file` | object {url, content_type, file_name, file_size} | Konfiguracija za nastavitev končne točke sklepanja. |
| `debug_dataset` | object {url, content_type, file_name, file_size} \| null | Izhod za odpravljanje napak, na voljo ob omogočenem `debug_dataset`: arhiv, ki ga je mogoče prenesti in vsebuje učne podatke po predobdelavi. |

## Related models

- [LTX 2.3 Trainer (V2) - Audio Inpainting](https://doitong.com/developers/models/ltx23-trainer-v2/audio-inpaint.md) (`ltx23-trainer-v2/audio-inpaint`): $0.0023 / step
- [LTX 2.3 Trainer (V2) - Audio+Video Reference IC-LoRA](https://doitong.com/developers/models/ltx23-trainer-v2/ic-lora/av2av.md) (`ltx23-trainer-v2/ic-lora/av2av`): $0.0068 / step
- [LTX 2.3 Trainer (V2) - Audio+Video Reference Transformation](https://doitong.com/developers/models/ltx23-trainer-v2/av2av.md) (`ltx23-trainer-v2/av2av`): $0.0068 / step
- [LTX 2.3 Trainer (V2) - Audio-to-Audio](https://doitong.com/developers/models/ltx23-trainer-v2/a2a.md) (`ltx23-trainer-v2/a2a`): $0.0023 / step
- [LTX 2.3 Trainer (V2) - Audio-to-Audio IC-LoRA](https://doitong.com/developers/models/ltx23-trainer-v2/ic-lora/a2a.md) (`ltx23-trainer-v2/ic-lora/a2a`): $0.0023 / step
- [LTX 2.3 Trainer (V2) - Audio-to-Video](https://doitong.com/developers/models/ltx23-trainer-v2/a2v.md) (`ltx23-trainer-v2/a2v`): $0.006 / step
- [LTX 2.3 Trainer (V2) - Backward Audio Extension](https://doitong.com/developers/models/ltx23-trainer-v2/audio-extend-suffix.md) (`ltx23-trainer-v2/audio-extend-suffix`): $0.0023 / step
- [LTX 2.3 Trainer (V2) - Forward Audio Extension](https://doitong.com/developers/models/ltx23-trainer-v2/audio-extend-prefix.md) (`ltx23-trainer-v2/audio-extend-prefix`): $0.0022 / step

## Links

- HTML page: https://doitong.com/developers/models/ltx23-trainer-v2/extend-suffix
- All models and guides for AI agents: https://doitong.com/developers/llms.txt
- OpenAPI 3.1 with this model's typed request body: https://doitong.com/developers/openapi.json?model=ltx23-trainer-v2/extend-suffix
- Errors: https://doitong.com/developers/docs/errors
- Model details as JSON: https://api.doitong.com/v1/models/ltx23-trainer-v2/extend-suffix