# MiniMax H3 Reference to Video LoRA Trainer API

> LoRA trainer para sa reference-conditioned video, na may warm start mula sa naunang LoRA

Nagsasanay ng LoRA sa zip ng mga video clip ng subject upang magabayan ng reference input ang pagbuo ng video. Itinatakda ng `reference_conditioning_p` kung gaano kadalas ginagamit ang mga reference habang nagsasanay, at ipinagpapatuloy ng `resume_from_lora_url` ang pagsasanay mula sa naunang LoRA. Maaari ring i-configure ang `rank`, `number_of_steps`, `number_of_frames`, `frame_rate` at `aspect_ratio`. Nagbabalik ng weights at config file.

- Model id: `minimax/h3/ref2va/trainer`
- Made by: MiniMax
- Category: Training
- Output: file
- Price: $0.015 / step (cheapest run $30.00 (30000 API credits))
- Run: `POST https://api.doitong.com/v1/run/minimax/h3/ref2va/trainer`
- Quote: `POST https://api.doitong.com/v1/quote/minimax/h3/ref2va/trainer`
- Docs page: https://doitong.com/fil/developers/models/minimax/h3/ref2va/trainer

## Pricing

Price: **$0.015 / step**. Billed per training step (`number_of_steps`).

The full example input below costs $30.00 = 30000 API credits (2000 steps × 15 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/minimax/h3/ref2va/trainer` 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 para sa ZIP ng subject training video; tinatanggihan ang larawan-only dataset at mixed larawan/video training archive, kaya dapat reference sidecar ang mga larawan. Ang mga opsyonal na karagdagang file ay kapareho ng base name ng bawat clip: caption gaya ng `clip01.txt` at mga nakaayos na reference mula `ref_1` hanggang `ref_4`, na pinapangalanang `clip01.ref_1.<ext>` hanggang `clip01.ref_4.<ext>`; kung walang reference sidecar, frame mula sa clip ang nagsisilbing sarili nitong reference. Tinutukoy ng `<ext>` ang reference modality: larawan (.png/.jpg/.jpeg), audio (.wav/.mp3/.flac/.m4a/.ogg/.aac), o videos (.mp4/.mov/.avi/.mkv/.webm), na nagiging audio reference din ang mga soundtrack ng mga ito. Sinusuportahan ng training clip ang .mp4, .mov, .avi, at .mkv; maghangad ng hindi bababa sa 10 clip, at mas mainam ang higit pa. |
| `rank` |  | enum (integer) | no | `32` | `8`, `16`, `32`, `64`, `128` | LoRA adaptation rank, na ipinagpapalit ang mas mataas na memory consumption para sa mas malaking capacity habang tumataas ang setting. |
| `number_of_steps` |  | integer | no | `2000` | 1 – 15,000 | Kabuuang bilang ng training step. |
| `learning_rate` |  | number | no | `0.0002` | 0.000001 – 1 | Optimization learning rate, na nagbabalanse sa posibleng pagbilis ng training at panganib ng overfitting kapag tinaasan. |
| `number_of_frames` |  | integer | no | `73` | 22 – 124 | Bilang ng frame para sa bawat training sample; kapag hinati sa 17, dapat may remainder na 5. Kabilang sa mga halimbawa ang 22, 39, 56, 73, 90, 107, at 124. |
| `frame_rate` |  | integer | no | `24` | 8 – 60 | Target frame rate ng video, sinusukat sa frames per segundo. |
| `resolution` |  | enum (string) | no | `medium` | `low`, `medium`, `high` | Training resolution. Kapag tinaasan ang resolution, tumataas ang memory requirements. |
| `aspect_ratio` |  | enum (string) | no | `16:9` | `21:9`, `16:9`, `4:3`, `1:1`, `3:4`, `9:16` | Piliin ang training aspect ratio mula sa parehong set na available sa MiniMax H3 API. |
| `trigger_phrase` |  | string | no | `` |  | LoRA style activation phrase, na ipinapasok sa simula ng mga caption habang nagta-training. |
| `reference_conditioning_p` |  | number | no | `0.9` | 0 – 1 | Itakda kung gaano kadalas gumagamit ang training ng reference-image conditioning kumpara sa unconditioned text-to-video-audio training. Napupunta ang natitirang probability sa unconditioned objective upang mapanatiling stable ang prompt-only pagbuo. |
| `resume_from_lora_url` |  | string \| null | no |  |  | URL para sa warm-start training gamit ang .safetensors LoRA na dati nang ginawa ng trainer na ito, na sinusundan ng bilang ng karagdagang training step na number_of_steps. Walang ibinabalik maliban sa adapter weights: muling nagsisimula ang optimization at learning-rate scheduling. |
| `auto_scale_input` |  | boolean | no | `false` |  | Kung nagtutugma ang target fps at frame count sa pamamagitan ng awtomatikong video scaling; ino-enable ng true ang adjustment na ito. |
| `split_input_into_scenes` |  | boolean | no | `true` |  | I-enable ang scene splitting para sa mga video na mas mahaba sa threshold ng tagal, at itapon ang reference sidecar mula sa mga clip na hinati. |
| `split_input_duration_threshold` |  | number | no | `30` | 1 – 60 | Trigger ng paghahati ng eksena na ipinapahayag bilang duration sa segundo: ang mga video lamang na mas mahaba sa threshold ang hinahati. |
| `debug_dataset` |  | boolean | no | `false` |  | Opsyon upang mag-download ng archive para suriin ang preprocessing bago simulan ang buong training. Kapag enabled ito, ibinabalot ng trainer ang processed dataset upang mano-manong masuri ang mga caption, reference larawan, video, at larawan kung tama. |
| `strict_dataset` |  | boolean | no | `false` |  | Mahigpit na paghawak sa mga preprocessing problem na kung hindi ay ila-log at lulutasin gamit ang dokumentadong fallback. Kapag enabled, tinatanggihan ang caption na nangangailangan ng blank prompt, nawawalang reference sidecar, at video na hindi mabasa ang audio sa halip na payagan ang mga fallback na iyon. |

## 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/minimax/h3/ref2va/trainer?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/minimax/h3/ref2va/trainer?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/minimax/h3/ref2va/trainer", 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": "minimax/h3/ref2va/trainer",
      "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": "minimax/h3/ref2va/trainer",
  "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": 30000,
    "usd": "30.00",
    "charged": 30000,
    "refunded": 0
  },
  "input": {
    "training_data_url": "https://example.com/your-file.zip"
  },
  "output": {
    "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 |
|---|---|---|
| `lora_file` | object {url, content_type, file_name, file_size} | URL para sa trained LoRA weights sa .safetensors format. |
| `config_file` | object {url, content_type, file_name, file_size} | Configuration para sa setup ng inference endpoint. |
| `debug_dataset` | object {url, content_type, file_name, file_size} \| null | Available lamang para sa mga kahilingang may naka-enable na `debug_dataset`; naglalaman ang mada-download na archive na ito ng training data pagkatapos ng preprocessing. |

## Related models

- [MiniMax H3 First Last Frame LoRA Trainer](https://doitong.com/developers/models/minimax/h3/flf2v/trainer.md) (`minimax/h3/flf2v/trainer`): $0.01 / step
- [MiniMax H3 Image to Video Audio LoRA Trainer](https://doitong.com/developers/models/minimax/h3/i2v/trainer.md) (`minimax/h3/i2v/trainer`): $0.01 / step
- [MiniMax H3 Text to Video LoRA Trainer](https://doitong.com/developers/models/minimax/h3/t2v/trainer.md) (`minimax/h3/t2v/trainer`): $0.005 / step
- [ERNIE-Image Trainer](https://doitong.com/developers/models/ernie-image-trainer.md) (`ernie-image-trainer`): $0.0012 / step
- [FLUX 2 \[klein\] 9b Base Trainer](https://doitong.com/developers/models/flux-2-klein-9b-base-trainer.md) (`flux-2-klein-9b-base-trainer`): $0.0043 / step
- [Flux 2 Klein 9B Base Trainer](https://doitong.com/developers/models/flux-2-klein-9b-base-trainer/edit.md) (`flux-2-klein-9b-base-trainer/edit`): $0.0139 / step
- [FLUX 2 Trainer](https://doitong.com/developers/models/flux-2-trainer.md) (`flux-2-trainer`): $0.0064 / step
- [FLUX 2 Trainer Edit](https://doitong.com/developers/models/flux-2-trainer/edit.md) (`flux-2-trainer/edit`): $0.0389 / step

## Links

- HTML page: https://doitong.com/developers/models/minimax/h3/ref2va/trainer
- 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=minimax/h3/ref2va/trainer
- Errors: https://doitong.com/developers/docs/errors
- Model details as JSON: https://api.doitong.com/v1/models/minimax/h3/ref2va/trainer