# LTX 2.3 Trainer (V2) - Audio+Video Reference Transformation API

> Furma ina LoRA che mappa il video e l'audio d'in clip da referenza ad in clip da destinaziun accumpagnà, cun las duas pistas generadas ensemen

Adattescha ina LoRA a clips accopladads, nua che mintga video da referenza cun ses audio vegn collià ad in clip da mira. Furnescha in zip via `training_data_url`, cun descripziuns .txt facultativas. La separaziun da scenas è deactivada per standard perquai ch'ella disturbass il sincronissem da las parejas entschatta/finala; furnescha pia clips gia tagliads. Rank po esser da 8 fin 128, pass da 100 fin 10000, e fin a 2 entradas da validaziun èn permessas.

- Model id: `ltx23-trainer-v2/av2av`
- Made by: Lightricks
- Category: Training
- Output: file
- Price: $0.0068 / step (cheapest run $13.60 (13600 API credits))
- Run: `POST https://api.doitong.com/v1/run/ltx23-trainer-v2/av2av`
- Quote: `POST https://api.doitong.com/v1/quote/ltx23-trainer-v2/av2av`
- Docs page: https://doitong.com/rm/developers/models/ltx23-trainer-v2/av2av

## Pricing

Price: **$0.0068 / step**. Billed per training step (`number_of_steps`).

The full example input below costs $13.60 = 13600 API credits (2000 steps × 6.8 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/av2av` 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 per in `.zip` cun datas da scolaziun, organisadas tenor il modus da scolaziun tschernì. Mintga datoteca da media po avair ina legenda `.txt` cun il medem num da basa. |
| `rank` |  | enum (integer) | no | `32` | `8`, `16`, `32`, `64`, `128` | Rank da l'adattaziun LoRA, che baratta dapli consum da memoria cunter dapli capacitad sch'il parameter crescha. |
| `number_of_steps` |  | integer | no | `2000` | 100 – 10,000 | Dumber total da pass da scolaziun. |
| `learning_rate` |  | number | no | `0.0002` | 0.000001 – 1 | Rate d'emprender per l'optimizaziun. |
| `number_of_frames` |  | integer | no | `89` | 9 – 121 | Dumber da frames per mintga exempel da scolaziun. Divisà il dumber cun 8 sto dar in rest da 1; exempels èn 1, 9, 17, 25, 33, 41, 49, 57, 65, 73, 81, 89, 97. |
| `frame_rate` |  | integer | no | `24` | 8 – 60 | Ritmus da frames da la video da scolaziun en frames per secunda. Il ritmus nativ per LTX-2.3 è 24. |
| `resolution` |  | enum (string) | no | `medium` | `low`, `medium`, `high` | Resoluziun da scolaziun. Augmentar la resoluziun augmenta il basegn da memoria. |
| `aspect_ratio` |  | enum (string) | no | `1:1` | `16:9`, `1:1`, `9:16` | Proporziun da la scolaziun. |
| `trigger_phrase` |  | string | no | `` |  | Frasa d'activaziun dal stil LoRA, inserida a l'entschatta da las legendas durant la scolaziun. |
| `auto_scale_input` |  | boolean | no | `false` |  | Adattaziun automatica dal video per correspunder a FPS e dumber da frames dumandads sch'activada; datasets da maletgs restan senza midadas. |
| `split_input_into_scenes` |  | boolean | no | `false` |  | La partiziun da scenas è disactivada en av2av per mantegnair il pè _start/_end sincronisà. Ils clips ston vegnir partidas ordavant; la preparaziun dal pè target/referenza ignora la configuraziun da partiziun da scenas. |
| `split_input_duration_threshold` |  | number | no | `30` | 1 – 60 | Trigger da partiziun da scenas exprimì sco durada en secundas: mo videos pli lungs che la limita vegnan partidas. |
| `debug_dataset` |  | boolean | no | `false` |  | Modus d'inspecziun da preelavuraziun per controllar videos, maletgs e legendas avant ina scolaziun cumpletta. L'activaziun fa turnar al trainer il dataset da scolaziun elavurà en in archiv telechargiabel per controlla manuala. |
| `audio_normalize` |  | boolean | no | `true` |  | Normalisaziun dals pics da l'audio per mantegnair nivels consistents en l'entir dataset. |
| `audio_preserve_pitch` |  | boolean | no | `true` |  | Mantegniment da l'autezza durant l'adattaziun da l'audio a la lunghezza dal video empè da al tagliar. |
| `validation` |  | array of object {prompt, reference_video_url, reference_audio_url} | no | `[]` | ≤ 2 items | Glista da prompts da validaziun che colla mintga prompt cun in video e audio da referenza. |
| `validation[].prompt` |  | string | yes |  |  | Prompt utilisà durant la validaziun. |
| `validation[].reference_video_url` |  | string (video URL) | yes |  | ≥ 1 characters | URL dal video da referenza utilisà per il conditioning da la generaziun audio e video. La pista audio dal video serva era sco referenza audio. |
| `validation[].reference_audio_url` |  | string (audio URL) \| null | no |  |  | Sustituescha facultativamain la pista audio da `reference_video_url` furnind ina URL per ina datoteca audio da referenza separada. |
| `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.` |  | Prompt negativ applitgà durant la validaziun, che utilisescha in process d'antevista en ina singula fasa. Perquai che quel approximà l'inferenza da producziun distillada en duas fasas, sa differenzian sia guida e qualitad d'antevista dals resultats finals. |
| `validation_number_of_frames` |  | integer | no | `89` | 9 – 121 | Dumber da frames per mintga video da validaziun. |
| `validation_frame_rate` |  | integer | no | `24` | 8 – 60 | Ritmus da frames target per videos da validaziun, en frames per secunda; LTX-2.3 funcziunescha nativamain a 24. |
| `validation_resolution` |  | enum (string) | no | `high` | `low`, `medium`, `high` | Resoluziun utilisada durant la validaziun. |
| `validation_aspect_ratio` |  | enum (string) | no | `1:1` | `16:9`, `1:1`, `9:16` | Proporziun utilisada durant la validaziun. |
| `stg_scale` |  | number | no | `1` | 0 – 3 | Scala da Spatio-Temporal Guidance (STG), cun 1.0 recumandà e 0.0 che deactivescha STG. |
| `reference_downscale_factor` |  | integer | no | `1` | 1 – 8 | Factor da reducziun per crear in proxy da controlla da video da referenza grossier, a bassa resoluziun, avant l'encoding, mantegnend l'output a resoluziun cumpletta. Las duas dimensiuns da referenza ston esser divisiblas cun 32 avant e suenter la division tras il factor; il standard 1 mantegn la resoluziun originala. |
| `reference_temporal_scale_factor` |  | integer | no | `1` | 1 – 8 | Factor da reducziun dal ritmus da frames dal video da referenza, cun 1 sco standard per mantegnair fps. Il downsampling temporal tras quest factor reducescha fps avant che codar il video. |

## 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/av2av?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/av2av?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/av2av", 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/av2av",
      "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/av2av",
  "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": 13600,
    "usd": "13.60",
    "charged": 13600,
    "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 | Antevista video cumaschada da la validaziun, sch'ella è disponibla per endpoints che produceschan video. |
| `audio` | object {url, content_type, file_name, file_size} \| null | Audio cumaschà da la validaziun, furnì sco antevista sch'el è disponibel per questas rutas mo-audio: /audio-inpaint, /audio-extend-suffix, /audio-extend-prefix, /t2a, /a2a e /v2a. |
| `lora_file` | object {url, content_type, file_name, file_size} | URL per las ponderaziuns LoRA scoladas en format .safetensors. |
| `config_file` | object {url, content_type, file_name, file_size} | Configuraziun per l'installaziun da l'endpoint d'inferenza. |
| `debug_dataset` | object {url, content_type, file_name, file_size} \| null | Output da debug disponibel cun `debug_dataset` activà: in archiv telechargiabel che cuntegna las datas da scolaziun suenter la preelavuraziun. |

## 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-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) - Backward Video Extension](https://doitong.com/developers/models/ltx23-trainer-v2/extend-suffix.md) (`ltx23-trainer-v2/extend-suffix`): $0.0061 / 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/av2av
- 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/av2av
- Errors: https://doitong.com/developers/docs/errors
- Model details as JSON: https://api.doitong.com/v1/models/ltx23-trainer-v2/av2av