# AIVidLab API — Complete Reference for AI Agents
# Last updated: 2026-07-03
# Base URL: https://api.aividlab.shop
# Authentication: Bearer token in Authorization header
# Get API key: /apikey command in @AIVidLab_bot on Telegram
# Dashboard: https://api.aividlab.shop/dashboard
# Token rate: 1 token = RM0.10 (Malaysian Ringgit)

## AUTHENTICATION

All requests require:
```
Authorization: Bearer YOUR_KEY
```

Get your key:
1. Open @AIVidLab_bot on Telegram
2. Send /apikey
3. Copy the key

## PRICING

1 token = RM0.10
Topup packs: RM5 (50 tok), RM10 (100 tok), RM30 (300 tok), RM50 (500 tok +5% bonus), RM100 (1000 tok +5% bonus)

---

## VIDEO MODELS

### Grok Video
- Model ID: `grok-video`
- Modes: text-to-video, image-to-video
- Resolutions: 720p
- Aspect Ratios: 16:9, 9:16
- Pricing:
  - 10s = 7 tokens (RM0.70)
  - 20s = 12 tokens (RM1.20)
  - 30s = 17 tokens (RM1.70)

```bash
# Text-to-video
curl -s -X POST https://api.aividlab.shop/v1/video/generate \
-H "Authorization: Bearer YOUR_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "grok-video",
    "prompt": "A cat playing piano in a jazz bar",
    "duration": 10,
    "aspect_ratio": "16:9"
  }'

# Image-to-video (pass image_url for i2v)
curl -s -X POST https://api.aividlab.shop/v1/video/generate \
-H "Authorization: Bearer YOUR_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "grok-video",
    "prompt": "The subject walks forward confidently",
    "image_url": "https://example.com/photo.jpg",
    "duration": 10,
    "aspect_ratio": "9:16"
  }'
```

---

### Omni Flash
- Model ID: `omni-flash`
- Modes: text-to-video, image-to-video
- Resolutions: 720p, 1080p, 4k
- Durations: 6s, 8s, 10s
- Aspect Ratios: 16:9, 9:16, 1:1
- Pricing: 16 tokens per generation (RM1.60) — flat rate regardless of duration/resolution

```bash
# Text-to-video
curl -s -X POST https://api.aividlab.shop/v1/video/generate \
-H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "omni-flash",
    "prompt": "Cinematic shot of a sunset over mountains",
    "duration": 10,
    "resolution": "720p",
    "aspect_ratio": "16:9"
  }'

# Image-to-video
curl -s -X POST https://api.aividlab.shop/v1/video/generate \
-H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "omni-flash",
    "prompt": "The subject walks forward confidently",
    "image_url": "https://example.com/photo.jpg",
    "duration": 10,
    "resolution": "720p",
    "aspect_ratio": "16:9"
  }'
```

---

### Seedance 2.0 Fast
- Model ID: `seedance-2-fast`
- Modes: text-to-video, image-to-video, multimodal
- Resolutions: 480p, 720p, 1080p, 2k, 4k
- Durations: 5s, 10s, 15s
- Aspect Ratios: 16:9, 9:16, 1:1
- Audio: included automatically
- Pricing (tokens):

| Resolution | 5s | 10s | 15s |
|---|---|---|---|
| 480p | 23 (RM2.30) | 40 (RM4.00) | 56 (RM5.60) |
| 720p | 40 (RM4.00) | 73 (RM7.30) | 110 (RM11.00) |
| 1080p | 48 (RM4.80) | 92 (RM9.20) | 137 (RM13.70) |
| 2k | 52 (RM5.20) | 101 (RM10.10) | 151 (RM15.10) |
| 4k | 58 (RM5.80) | 114 (RM11.40) | 172 (RM17.20) |

```bash
curl -s -X POST https://api.aividlab.shop/v1/video/generate \
-H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "seedance-2-fast",
    "prompt": "A dancer performing contemporary dance in a studio",
    "duration": 5,
    "resolution": "720p",
    "aspect_ratio": "16:9"
  }'
```

---

### Seedance 2.0 Standard
- Model ID: `seedance-2-standard`
- Modes: text-to-video, image-to-video, multimodal
- Resolutions: 480p, 720p, 1080p, native1080p, native4k
- Durations: 5s, 10s, 15s
- Aspect Ratios: 16:9, 9:16, 1:1
- Audio: included automatically
- Note: native1080p and native4k are TRUE native rendering (not upscaled)
- Pricing (tokens):

| Resolution | 5s | 10s | 15s |
|---|---|---|---|
| 480p | 27 (RM2.70) | 48 (RM4.80) | 69 (RM6.90) |
| 720p | 48 (RM4.80) | 92 (RM9.20) | 137 (RM13.70) |
| 1080p | 56 (RM5.60) | 110 (RM11.00) | 165 (RM16.50) |
| native1080p | 137 (RM13.70) | 229 (RM22.90) | 343 (RM34.30) |
| native4k | 229 (RM22.90) | 457 (RM45.70) | 685 (RM68.50) |

```bash
curl -s -X POST https://api.aividlab.shop/v1/video/generate \
-H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "seedance-2-standard",
    "prompt": "Aerial drone shot of a tropical island at golden hour",
    "duration": 10,
    "resolution": "native1080p",
    "aspect_ratio": "16:9"
  }'
```

---

### Wan 2.7
- Model ID: `wan-2.7`
- Modes: t2v, i2v, spicy_i2v, reference_to_video, video_extend, video_edit
- Resolutions: 720P, 1080P
- Durations: 2–15s (mode-dependent; ref2v/edit max 10s)
- Aspect Ratios: 16:9, 9:16, 1:1, 4:3, 3:4
- Pricing (tokens):

| Mode | Resolution | 5s | 10s | 15s |
|---|---|---|---|---|
| t2v / i2v / extend | 720P | 20 (RM1.97) | 37 (RM3.44) | 53 (RM4.91) |
| t2v / i2v / extend | 1080P | 30 (RM2.52+fee) | 56 (RM5.54) | 81 (RM8.06) |
| spicy_i2v | 720P | 32 (RM3.23) | 60 (RM5.96) | 87 (RM8.69) |
| spicy_i2v | 1080P | 46 (RM4.70) | 89 (RM8.90) | 132 (RM13.10) |
| ref2v / video_edit | 720P | 20 (RM1.97) 5s out | 27 (RM2.67) 10s out | — |
| ref2v / video_edit | 1080P | 30 (RM2.52+fee) 5s out | 43 (RM4.20+fee) 10s out | — |

- Additional params: `prompt_extend` (bool), `seed` (0-2147483647), `negative_prompt`, `audio_url`
- video_edit requires `input_duration` for billing accuracy
- reference_to_video accepts `video_urls` (max 5) + `image_urls` (max 5)

```bash
curl -s -X POST https://api.aividlab.shop/v1/video/generate \
  -H "Authorization: Bearer ***" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "wan-2.7",
    "mode": "t2v",
    "prompt": "Product showcase spinning on a white background",
    "duration": 5,
    "resolution": "720P",
    "aspect_ratio": "16:9"
  }'
```

---

### Sora 2
- Model ID: `sora-2`
- Modes: text-to-video
- Resolutions: 1080p
- Durations: 4s, 8s, 12s, 16s, 20s
- Aspect Ratios: 16:9, 9:16, 1:1
- Pricing:
  - 4s = 28 tokens (RM2.80)
  - 8s = 46 tokens (RM4.60)
  - 12s = 65 tokens (RM6.50)
  - 16s = 84 tokens (RM8.40)
  - 20s = 103 tokens (RM10.30)

```bash
curl -s -X POST https://api.aividlab.shop/v1/video/generate \
-H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "sora-2",
    "prompt": "A cinematic tracking shot through a futuristic city at night",
    "duration": 8,
    "aspect_ratio": "16:9"
  }'
```

---

### HappyHorse 1.0
- Model ID: `happyhorse-1.0`
- Modes: text-to-video (t2v), image-to-video (i2v)
- Resolutions: 720p, 1080p
- Durations: 3s, 4s, 5s, 6s, 7s, 8s, 9s, 10s, 11s, 12s, 13s, 14s, 15s
- Aspect Ratios: 16:9, 9:16, 1:1, 4:3, 3:4
- Pricing (720p):
  - 3s = 22 tokens (RM2.20)
  - 5s = 34 tokens (RM3.40)
  - 10s = 64 tokens (RM6.40)
  - 15s = 94 tokens (RM9.40)
- Pricing (1080p):
  - 3s = 26 tokens (RM2.60)
  - 5s = 40 tokens (RM4.00)
  - 10s = 76 tokens (RM7.60)
  - 15s = 112 tokens (RM11.20)

```bash
curl -s -X POST https://api.aividlab.shop/v1/video/generate \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "happyhorse-1.0",
    "prompt": "A playful kitten chasing butterflies in a garden",
    "duration": 5,
    "resolution": "720p",
    "aspect_ratio": "16:9"
  }'
```

---

### HappyHorse 1.1
- Model ID: `happyhorse-1.1`
- Modes: text-to-video (t2v), image-to-video (i2v)
- Resolutions: 720p, 1080p
- Durations: 3s, 4s, 5s, 6s, 7s, 8s, 9s, 10s, 11s, 12s, 13s, 14s, 15s
- Aspect Ratios: 16:9, 9:16, 1:1, 4:3, 3:4
- Pricing: Same as 1.0 — quality upgrade, no price change
- Note: HappyHorse 1.1 is a quality improvement over 1.0 at the same price point

```bash
curl -s -X POST https://api.aividlab.shop/v1/video/generate \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "happyhorse-1.1",
    "prompt": "Cinematic slow-motion of a horse galloping through mist at sunrise",
    "duration": 10,
    "resolution": "1080p",
    "aspect_ratio": "16:9"
  }'
```

---

### Kling Standard
- Model ID: `kling-std`
- Modes: text-to-video (t2v), image-to-video (i2v), reference-to-video (ref2v), video-edit, motion-control
- Resolutions: 1080p
- Durations: 3-15s (any integer)
- Aspect Ratios: 16:9, 9:16, 1:1
- Pricing (tokens) — UPDATED 2026-07-05:

| Mode | Duration | Tokens |
|---|---|---|
| t2v | 3s | 11 (RM1.10) |
| t2v | 5s | 19 (RM1.90) |
| t2v | 10s | 38 (RM3.80) |
| t2v | 15s | 52 (RM5.20) |
| i2v | 10s | 48 (RM4.80) |
| ref2v | 10s | 48 (RM4.80) |
| video_edit | flat | 50 (RM5.00) |
| motion | flat | 57 (RM5.70) |

```bash
# Text-to-video
curl -s -X POST https://api.aividlab.shop/v1/video/generate \
-H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "kling-std",
    "mode": "t2v",
    "prompt": "A samurai drawing his katana in the rain, cinematic lighting",
    "duration": 5,
    "aspect_ratio": "16:9",
    "sound": true
  }'
```

---

### Kling Pro
- Model ID: `kling-pro`
- Modes: text-to-video (t2v), image-to-video (i2v), reference-to-video (ref2v), video-edit, motion-control
- Resolutions: 1080p
- Durations: 3-15s (any integer)
- Aspect Ratios: 16:9, 9:16, 1:1
- Native sound sync: automatically enabled
- Pricing (tokens) — UPDATED 2026-07-05:

| Mode | Duration | Tokens |
|---|---|---|
| t2v | 3s | 14 (RM1.40) |
| t2v | 5s | 24 (RM2.40) |
| t2v | 10s | 48 (RM4.80) |
| t2v | 15s | 66 (RM6.60) |
| i2v | 10s | 62 (RM6.20) |
| ref2v | 10s | 62 (RM6.20) |
| video_edit | flat | 31 (RM3.10) |
| motion | flat | 50 (RM5.00) |

```bash
# Image-to-video
curl -s -X POST https://api.aividlab.shop/v1/video/generate \
-H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "kling-pro",
    "mode": "i2v",
    "prompt": "The woman walks forward, hair flowing in wind",
    "image_url": "https://your-image-url.jpg",
    "last_image_url": null,
    "duration": 5,
    "sound": true
  }'

# Reference-to-video
curl -s -X POST https://api.aividlab.shop/v1/video/generate \
-H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "kling-pro",
    "mode": "ref2v",
    "prompt": "The character performs a spin move",
    "image_urls": ["https://ref-image-1.jpg"],
    "video_url": "https://ref-video.mp4",
    "duration": 5,
    "aspect_ratio": "16:9"
  }'

# Video edit
curl -s -X POST https://api.aividlab.shop/v1/video/generate \
-H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "kling-std",
    "mode": "video_edit",
    "prompt": "Change the time to night, add rain",
    "video_url": "https://your-video.mp4"
  }'

# Motion control
curl -s -X POST https://api.aividlab.shop/v1/video/generate \
-H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "kling-pro",
    "mode": "motion",
    "image_url": "https://subject-image.jpg",
    "video_url": "https://reference-motion.mp4",
    "orientation": "video"
  }'
```

Orientation options for motion mode:
- `image` — Subject Mode (≤10s): follow subject image orientation
- `video` — Video Mode (≤30s): follow reference video motion

---

### Kling 4K Ultra HD
- Model ID: `kling-4k`
- Modes: text-to-video (t2v), image-to-video (i2v), reference-to-video (ref2v)
- Resolutions: 4K
- Durations: 3-15s (any integer)
- Aspect Ratios: 16:9, 9:16, 1:1
- Pricing (tokens) — UPDATED 2026-07-05:

| Mode | Duration | Tokens |
|---|---|---|
| t2v | 3s | 56 (RM5.60) |
| t2v | 5s | 94 (RM9.40) |
| t2v | 10s | 188 (RM18.80) |
| t2v | 15s | 276 (RM27.60) |
| i2v | 10s | 188 (RM18.80) |
| ref2v | 10s | 188 (RM18.80) |

```bash
curl -s -X POST https://api.aividlab.shop/v1/video/generate \
-H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "kling-4k",
    "mode": "t2v",
    "prompt": "Ultra-detailed aerial shot of a luxury resort at sunset",
    "duration": 10,
    "aspect_ratio": "16:9"
  }'
```

---

### Veo 3.1 Stable
- Model ID: `veo-3.1-stable`
- Modes: text-to-video, image-to-video
- Resolutions: 720p, 1080p
- Durations: 4s, 6s, 8s
- Aspect Ratios: 16:9, 9:16, 1:1
- Pricing (tokens):

| Resolution | 4s | 6s | 8s |
|---|---|---|---|
| 720p | 12 (RM1.20) | 16 (RM1.60) | 20 (RM2.00) |
| 1080p | 17 (RM1.70) | 23 (RM2.30) | 29 (RM2.90) |

```bash
curl -s -X POST https://api.aividlab.shop/v1/video/generate \
-H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "veo-3.1-stable",
    "prompt": "A beautiful sunset over the ocean, cinematic drone shot",
    "duration": 6,
    "resolution": "1080p",
    "aspect_ratio": "16:9"
  }'
```

---

### Veo 3.1 Pro
- Model ID: `veo-3.1-pro`
- Modes: text-to-video, image-to-video
- Resolutions: 720p, 1080p, 4K
- Duration: auto (not configurable)
- Aspect Ratios: 16:9, 9:16, 1:1
- Pricing (per-resolution, flat rate):

| Resolution | Cost |
|---|---|
| 720p | 9 tokens (RM0.90) |
| 1080p | 11 tokens (RM1.10) |
| 4K | 14 tokens (RM1.40) |

```bash
curl -s -X POST https://api.aividlab.shop/v1/video/generate \
-H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "veo-3.1-pro",
    "prompt": "Cinematic product showcase with dynamic camera movement",
    "resolution": "1080p",
    "aspect_ratio": "9:16"
  }'
```

---

### Veo 3.1 Fast
- Model ID: `veo-3.1-fast`
- Modes: text-to-video, image-to-video
- Resolutions: 720p, 1080p, 4K (same price)
- Duration: auto (not configurable)
- Aspect Ratios: 16:9, 9:16, 1:1
- Pricing: 12 tokens flat (RM1.20) — all resolutions same price
- Note: Fastest generation time of all Veo models

```bash
curl -s -X POST https://api.aividlab.shop/v1/video/generate \
-H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "veo-3.1-fast",
    "prompt": "Quick product video with zoom effects",
    "resolution": "4k",
    "aspect_ratio": "16:9"
  }'
```

---

## IMAGE MODELS

### GPT Image 2
- Model ID: `gpt-image-2`
- Modes: text-to-image, image-to-image (pass `image_urls` for edit)
- Pricing: 4 tokens per image (RM0.40)
- Aspect Ratios: 1:1, 16:9, 9:16

```bash
# Text-to-image
curl -s -X POST https://api.aividlab.shop/v1/image/generate \
-H "Authorization: Bearer YOUR_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gpt-image-2",
    "prompt": "A modern minimalist logo for a tech startup",
    "aspect_ratio": "1:1"
  }'

# Image-to-image (edit mode — pass image_urls)
curl -s -X POST https://api.aividlab.shop/v1/image/generate \
-H "Authorization: Bearer YOUR_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gpt-image-2",
    "prompt": "Change the background to a tropical beach sunset",
    "image_urls": ["https://example.com/original.jpg"],
    "aspect_ratio": "16:9"
  }'
```

---

### GPT Image 1.5
- Model ID: `gpt-image-1-5`
- Modes: text-to-image, image-to-image (pass `image_urls` for edit)
- Pricing: 4 tokens per image (RM0.40) — cheapest, fast, sharp text rendering
- Aspect Ratios: 1:1, 16:9, 9:16

```bash
curl -s -X POST https://api.aividlab.shop/v1/image/generate \
-H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gpt-image-1-5",
    "prompt": "A bold typographic poster that says SALE NOW",
    "aspect_ratio": "16:9"
  }'
```

---

### Nano Banana
- Model ID: `nano-banana`
- Modes: text-to-image, image-to-image (pass `image_urls` for edit)
- Pricing: 2 tokens per image (RM0.20) — cheapest image model
- Aspect Ratios: 1:1, 16:9, 9:16

```bash
curl -s -X POST https://api.aividlab.shop/v1/image/generate \
-H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "nano-banana",
    "prompt": "A cute cartoon cat sitting on a couch",
    "aspect_ratio": "1:1"
  }'
```

---

### Nano Banana 2
- Model ID: `nano-banana-2`
- Modes: text-to-image, image-to-image (pass `image_urls` for edit)
- Pricing: 4 tokens per image (RM0.40) — improved quality over Nano Banana
- Aspect Ratios: 1:1, 16:9, 9:16

```bash
curl -s -X POST https://api.aividlab.shop/v1/image/generate \
-H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "nano-banana-2",
    "prompt": "Professional headshot of a business woman, studio lighting",
    "aspect_ratio": "1:1"
  }'
```

---

### Nano Banana Pro
- Model ID: `nano-banana-pro`
- Modes: text-to-image, image-to-image (pass `image_urls` for edit)
- Pricing: 6 tokens per image (RM0.60) — best image quality
- Aspect Ratios: 1:1, 16:9, 9:16

```bash
curl -s -X POST https://api.aividlab.shop/v1/image/generate \
-H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "nano-banana-pro",
    "prompt": "Hyperrealistic food photography, gourmet burger with dramatic lighting",
    "aspect_ratio": "16:9"
  }'
```

### Grok
- Model ID: `grok`
- Pricing: 2 tokens per image (RM0.20) — fast & budget friendly
- Aspect Ratios: 1:1, 16:9, 9:16
- Edit mode: pass `image_urls` (max 1 reference image)

```bash
curl -s -X POST https://api.aividlab.shop/v1/image/generate \
-H "Authorization: Bearer *** \
  -H "Content-Type: application/json" \
  -d '{
    "model": "grok",
    "prompt": "A neon-lit cyberpunk street at night, rain reflections",
    "aspect_ratio": "9:16"
  }'
```

### Grok Stable
- Model ID: `grok-stable`
- Pricing: 2 tokens per image (RM0.20) — official, more stable & consistent
- Aspect Ratios: 1:1, 16:9, 9:16
- Edit mode: pass `image_urls` (max 1 reference image)

```bash
curl -s -X POST https://api.aividlab.shop/v1/image/generate \
-H "Authorization: Bearer *** \
  -H "Content-Type: application/json" \
  -d '{
    "model": "grok-stable",
    "prompt": "Professional product shot of a perfume bottle, soft studio light",
    "aspect_ratio": "1:1"
  }'
```

### Grok Quality
- Model ID: `grok-quality`
- Pricing: 5 tokens per image (RM0.50) — high quality, sharper detail
- Aspect Ratios: 1:1, 16:9, 9:16
- Edit mode: pass `image_urls` (max 1 reference image)

```bash
curl -s -X POST https://api.aividlab.shop/v1/image/generate \
-H "Authorization: Bearer *** \
  -H "Content-Type: application/json" \
  -d '{
    "model": "grok-quality",
    "prompt": "Cinematic portrait, dramatic rim lighting, ultra detailed",
    "aspect_ratio": "16:9"
  }'
```

### Seedream 4.5
- Model ID: `seedream-v45`
- Pricing: 3 tokens per image (RM0.30)
- Aspect Ratios: 1:1, 16:9, 9:16, 4:3, 3:4
- Edit mode: pass `image_urls` (up to 10 reference images)

```bash
curl -s -X POST https://api.aividlab.shop/v1/image/generate \
  -H "Authorization: Bearer *** \
  -H "Content-Type: application/json" \
  -d '{
    "model": "seedream-v45",
    "prompt": "Cinematic wuxia movie poster, dramatic night rain, ultra detailed",
    "aspect_ratio": "9:16"
  }'
```

### Seedream 5 Lite
- Model ID: `seedream-v5lite`
- Pricing: 3 tokens per image (RM0.30)
- Aspect Ratios: 1:1, 16:9, 9:16, 4:3, 3:4
- Edit mode: pass `image_urls` (up to 10 reference images)

```bash
curl -s -X POST https://api.aividlab.shop/v1/image/generate \
  -H "Authorization: Bearer *** \
  -H "Content-Type: application/json" \
  -d '{
    "model": "seedream-v5lite",
    "prompt": "Modern flat illustration weather forecast chart, clean vector style",
    "aspect_ratio": "16:9"
  }'
```

### Seedream 5 Pro
- Model ID: `seedream-v5pro`
- Pricing: 5 tokens per image (RM0.50)
- Aspect Ratios: 1:1, 16:9, 9:16, 4:3, 3:4
- Edit mode: pass `image_urls` (up to 10 reference images)

```bash
curl -s -X POST https://api.aividlab.shop/v1/image/generate \
  -H "Authorization: Bearer *** \
  -H "Content-Type: application/json" \
  -d '{
    "model": "seedream-v5pro",
    "prompt": "High-end skincare e-commerce hero shot, glass serum bottle, soft studio lighting, ultra detailed",
    "aspect_ratio": "9:16"
  }'
```

### Jimeng 4.6
- Model ID: `jimeng46`
- Pricing: 3 tokens per image (RM0.30)
- Aspect Ratios: 1:1, 16:9, 9:16, 4:3, 3:4
- Edit mode: pass `image_urls` (up to 6 reference images recommended)

```bash
curl -s -X POST https://api.aividlab.shop/v1/image/generate \
  -H "Authorization: Bearer *** \
  -H "Content-Type: application/json" \
  -d '{
    "model": "jimeng46",
    "prompt": "Professional Japanese anime illustration, full-body portrait, premium quality",
    "aspect_ratio": "9:16"
  }'
```

---

### Flux Klein
- Model ID: `flux-klein`
- Pricing: 2 tokens per image (RM0.20)
- Aspect Ratios: 1:1, 16:9, 9:16, 4:3, 3:4, 2:3, 3:2
- Edit mode: pass `image_urls` (1 reference image)

```bash
curl -s -X POST https://api.aividlab.shop/v1/image/generate \
  -H "Authorization: Bearer *** \
  -H "Content-Type: application/json" \
  -d '{
    "model": "flux-klein",
    "prompt": "Photorealistic product shot, soft studio lighting, crisp text on label",
    "aspect_ratio": "9:16"
  }'
```

---

## MUSIC & AUDIO MODELS

### Suno V5.5
- Model ID: `suno-v5.5`
- Category: Song generation
- Cost: 9 tokens (RM0.90)
- Required params: `title`, `lyrics`, `tags`
- Endpoint: POST /v1/music/generate

```
curl -s -X POST https://api.aividlab.shop/v1/music/generate \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "suno-v5.5",
    "title": "Rindu Di Hati",
    "lyrics": "[Verse]\nMalam ini aku sendiri\n[Chorus]\nRindu ini membara",
    "tags": "pop melayu, sedih, slow ballad, female vocal"
  }'
```

### Suno V5
- Model ID: `suno-v5`
- Cost: 9 tokens (RM0.90)
- Same params as suno-v5.5

### Suno V4.5
- Model ID: `suno-v4.5`
- Cost: 9 tokens (RM0.90)
- Same params as suno-v5.5

### Mureka 8 Song
- Model ID: `mureka-v8-song`
- Cost: 6 tokens (RM0.60)
- Required: `lyrics`
- Optional: `prompt` (style), `vocal_id`, `reference_id`, `melody_id`, `n` (1-3)

```
curl -s -X POST https://api.aividlab.shop/v1/music/generate \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "mureka-v8-song",
    "lyrics": "[Verse]\nAku berjalan tanpa arah\n[Chorus]\nHati ini masih rindu",
    "prompt": "upbeat pop, female vocal, acoustic guitar"
  }'
```

### Mureka 7.6 Song
- Model ID: `mureka-v7.6-song`
- Cost: 5 tokens (RM0.50)
- Same params as mureka-v8-song

### Mureka O2 Song
- Model ID: `mureka-o2-song`
- Cost: 6 tokens (RM0.60)
- Same params as mureka-v8-song

### MiniMax Music 2.6
- Model ID: `minimax-music-2.6`
- Cost: 9 tokens (RM0.90)
- Optional: `prompt` (style), `lyrics`, `is_instrumental` (bool)

```
curl -s -X POST https://api.aividlab.shop/v1/music/generate \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "minimax-music-2.6",
    "prompt": "cinematic orchestral epic",
    "is_instrumental": true
  }'
```

### Mureka 8 BGM
- Model ID: `mureka-v8-bgm`
- Category: Instrumental/BGM (no vocals)
- Cost: 6 tokens (RM0.60)
- Required: `prompt` (style/mood)

```
curl -s -X POST https://api.aividlab.shop/v1/music/generate \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "mureka-v8-bgm",
    "prompt": "calm lo-fi hip hop, rainy day vibes"
  }'
```

### Mureka 7.6 BGM
- Model ID: `mureka-v7.6-bgm`
- Cost: 5 tokens (RM0.50)
- Same params as mureka-v8-bgm

### Suno Lyrics
- Model ID: `suno-lyrics`
- Category: AI lyrics generation
- Cost: 4 tokens (RM0.40)
- Required: `prompt` (theme/topic)
- Response: `lyrics` text field (NOT audio URL)

```
curl -s -X POST https://api.aividlab.shop/v1/music/generate \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "suno-lyrics",
    "prompt": "lagu cinta putus asa, gaya pop melayu"
  }'
```

### Mureka AI Lyrics
- Model ID: `mureka-lyrics`
- Cost: 4 tokens (RM0.40)
- Same params as suno-lyrics

### Mureka 8 Extend
- Model ID: `mureka-v8-extend`
- Category: Continue/extend existing song
- Cost: 8 tokens (RM0.80)
- Required: `file_url` (audio URL), `lyrics` (new section)
- Optional: `extend_at` (ms, default 8000), `extend_type` ("tail"/"head")

```
curl -s -X POST https://api.aividlab.shop/v1/music/generate \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "mureka-v8-extend",
    "file_url": "https://cdn.example.com/song.mp3",
    "lyrics": "[Bridge]\nDi sini aku menunggu",
    "extend_type": "tail"
  }'
```

### Mureka 7.6 Extend
- Model ID: `mureka-v7.6-extend`
- Cost: 5 tokens (RM0.50)
- Same params (no extend_type)

### MiniMax Cover
- Model ID: `minimax-cover`
- Category: AI song cover
- Cost: 9 tokens (RM0.90)
- Required: `file_url` (audio URL)
- Optional: `prompt` (style), `lyrics`

```
curl -s -X POST https://api.aividlab.shop/v1/music/generate \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "minimax-cover",
    "file_url": "https://cdn.example.com/reference.mp3",
    "prompt": "acoustic jazz version"
  }'
```

### MiniMax Cover Preprocess
- Model ID: `minimax-cover-pre`
- Cost: 9 tokens (RM0.90)
- Required: `file_url`
- Returns: `coverFeatureId` (valid 24h, use for cover generation)

### Mureka Vocal Clone
- Model ID: `mureka-vocal-clone`
- Category: Voice cloning
- Cost: 206 tokens (RM20.60)
- Required: `file_url` (clean vocal, 10s–5min, MP3/WAV)
- Returns: `vocalId` — reuse with mureka-v8-song

```
curl -s -X POST https://api.aividlab.shop/v1/music/generate \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "mureka-vocal-clone",
    "file_url": "https://cdn.example.com/vocal-sample.mp3"
  }'
```

---

## TEXT-TO-SPEECH (TTS)

Endpoint: POST /v1/tts/generate
Poll status: GET /v1/music/{task_id} (same as music)
Result: audio MP3 URL (expires 24h)

### Available TTS Models
| Model ID | Cost | Features |
|----------|------|----------|
| speech-2.8-hd | 6 tok (RM0.60) | Latest HD, 40+ languages, emotion, interjections |
| speech-2.6-hd | 8 tok (RM0.80) | HD with emotion and pitch control |
| speech-2.6-turbo | 6 tok (RM0.60) | Fast, low latency |
| speech-02-hd | 6 tok (RM0.60) | High-definition voice |
| speech-02-turbo | 5 tok (RM0.50) | Budget, fastest speed |

### TTS Parameters
- `model` (required): Model ID from table above
- `text` (required): 1–10,000 characters
- `voice_id` (optional): Wise_Woman, Gentle_Woman, Warm_Man, Deep_Man, Narrator, Cheerful_Girl, Calm_Boy, Storyteller, Broadcaster, Customer_Service (default: Wise_Woman)
- `speed` (optional): 0.5–2.0 (default: 1.0)
- `volume` (optional): 0.1–10.0 (default: 1.0)
- `pitch` (optional): -12 to +12 semitones (default: 0)
- `emotion` (optional): neutral, happy, sad, angry, fearful, disgusted, surprised (default: neutral)

```
curl -s -X POST https://api.aividlab.shop/v1/tts/generate \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "speech-2.8-hd",
    "text": "Selamat datang ke AIVidLab.",
    "voice_id": "Warm_Man",
    "speed": 1.0,
    "emotion": "happy"
  }'
```

---

## ENDPOINTS REFERENCE

### Check Balance
```
GET /v1/balance
```
Response: `{"available": 150, "reserved": 0, "total_topup": 200, "total_spent": 50}`

### List Models
```
GET /v1/models
```
Returns all models with capabilities and pricing.

### Models Status (24H)
```
GET /v1/models/status
```
Returns real-time 24H health stats per model: success_rate, total generations, avg generation time, status (healthy/degraded/down).

Response:
```json
{
  "models": [
    {
      "model": "Super Grok",
      "status": "healthy",
      "success_rate": 96.7,
      "avg_generation_time": 95.2,
      "total_24h": 90
    }
  ]
}
```

### Models Status Timeline (144 segments)
```
GET /v1/models/status/timeline
```
Returns 144 segments per model (10-min buckets over 24 hours). Use this to render uptime-style status bars.

Response:
```json
{
  "period_start": 1782952980,
  "period_end": 1783039380,
  "bucket_size_seconds": 600,
  "models": [
    {
      "model": "Super Grok",
      "segments": ["green", "green", "grey", "red", ...],
      "success_rate": 96.7,
      "total": 90
    }
  ]
}
```
Segment colors: `green` = all success, `red` = all failed, `yellow` = mixed, `grey` = no data in that 10-min window.

### Generate Video
```
POST /v1/video/generate
```
Body:
- `model` (required): Model ID from list above
- `prompt` (required): Text description (max 5000 chars)
- `duration` (optional): Duration in seconds (default 10, range 3-30 depending on model)
- `resolution` (optional): e.g. "720p", "1080p", "4k", "native1080p", "native4k"
- `aspect_ratio` (optional): e.g. "16:9", "9:16", "1:1" (default "9:16")
- `image_url` (optional): Reference image URL for i2v mode
- `video_url` (optional): Reference video URL for multimodal/ref2v/video-edit/motion mode
- `mode` (optional): Kling mode — "t2v", "i2v", "ref2v", "video_edit", "motion". Auto-detected if omitted.
- `image_urls` (optional): Multiple reference image URLs (max 9, for multimodal/ref2v)
- `last_image_url` (optional): End frame image URL (Kling i2v mode)
- `orientation` (optional): Motion control orientation — "image" or "video" (Kling motion mode)
- `sound` (optional): Enable native sound sync (Kling Pro auto-enables)
- `negative_prompt` (optional): Describe what to avoid in the video (Kling)
- `cfg_scale` (optional): CFG scale for prompt adherence (Kling)
- `element_list` (optional): Kling element IDs for character/object consistency (max 3)

Response: `{"task_id": "123", "status": "submitted", "model": "grok-video", "cost_tokens": 7, "message": "Video generation submitted. Poll GET /v1/video/{task_id} for status."}`

### Check Video Status
```
GET /v1/video/{task_id}
```
Response: `{"task_id": "123", "status": "completed", "result_url": "https://...", "progress": 100, "error": null}`

Status values: `submitting`, `submitted`, `processing`, `completed`, `failed`

⚠️ IMPORTANT:
- Result URLs expire after 24 hours. Download and store on your own CDN immediately.
- Tokens are auto-refunded on failure (status becomes `refunded`).
- Recommended polling interval: every 5 seconds.
- Typical generation times: Veo Fast ~1min, Veo Pro ~2min, Kling ~2min, Seedance ~3min, Omni ~5min, Sora ~5min.
- Max polling duration: 10 minutes. If still `processing` after 10min, assume timeout (auto-refunded).

### Generate Image
```
POST /v1/image/generate
```
Body:
- `model` (required): Model ID
- `prompt` (required): Text description (max 5000 chars)
- `aspect_ratio` (optional): e.g. "1:1", "16:9", "9:16" (default "1:1")
- `image_urls` (optional): Reference image URLs for image-to-image/edit mode (max 10)

Response: `{"task_id": "456", "status": "submitted", "model": "gpt-image-2", "cost_tokens": 4, "message": "Image generation submitted. Poll GET /v1/image/{task_id} for status."}`

### Check Image Status
```
GET /v1/image/{task_id}
```
Response: `{"task_id": "456", "status": "completed", "result_url": "https://...", "error": null}`

### Generate Music/Audio
```
POST /v1/music/generate
```
Body: `{"model": "suno-v5.5", "title": "...", "lyrics": "...", "tags": "..."}`
Response: `{"task_id": "789", "status": "submitted", "model": "Suno V5.5", "cost_tokens": 9}`

See MUSIC & AUDIO MODELS section above for per-model required/optional params.

### Generate TTS
```
POST /v1/tts/generate
```
Body: `{"model": "speech-2.8-hd", "text": "Hello world", "voice_id": "Warm_Man", "emotion": "happy"}`
Response: `{"task_id": "790", "status": "submitted", "model": "Speech-2.8 HD", "cost_tokens": 6}`

### Check Music/TTS Status
```
GET /v1/music/{task_id}
```
Response: `{"task_id": "789", "status": "completed", "result_url": "https://...audio.mp3", "error": null}`

⚠️ Audio result URLs expire after 24 hours. Download immediately on completion.

### Upload File
```
POST /v1/upload
```
Upload an image or video file. Returns a public CDN URL for use as `image_url` or `video_url` in generation requests.

⚠️ Uploaded file URLs expire after 72 hours. Re-upload if needed after expiry.

- Content-Type: `multipart/form-data`
- Field: `file` (required)
- Max size: 10MB (images), 100MB (videos)
- Supported images: jpg, png, webp, gif
- Supported videos: mp4, webm, mov, avi

Response: `{"url": "https://...", "filename": "photo.jpg", "type": "image"}`

```bash
curl -s -X POST https://api.aividlab.shop/v1/upload \
-H "Authorization: Bearer YOUR_API_KEY" \
  -F "file=@/path/to/photo.jpg"
```

### Usage History
```
GET /v1/usage?limit=50&offset=0
```
Optional query params: `from` (YYYY-MM-DD), `to` (YYYY-MM-DD) — dates in MYT (UTC+8).

Response: `{"usage": [...], "total": 100, "stats": {"total_jobs": 100, "completed": 90, "failed": 10, "total_tokens_used": 1500}}`

### Generations List
```
GET /v1/generations?limit=20&offset=0&status=completed
```
List user's generation history (video + image combined). Includes prompt details.

Optional query params:
- `limit` (default 20, max 100)
- `offset` (default 0)
- `status` (optional filter: submitting, submitted, processing, completed, failed)

Response: `{"results": [...], "total": 50}`

### User Info
```
GET /v1/user/info
```
Get user profile info for dashboard display.

Response: `{"telegram_id": 123456, "username": "johndoe", "first_name": "John", "available": 150, "reserved": 0, "total_topup": 200, "total_spent": 50}`

### API Keys Management

List keys:
```
GET /v1/keys
```

Create key:
```
POST /v1/keys
```
Response: `{"key": "avl_...", "created": true}`

Revoke key:
```
DELETE /v1/keys/{key}
```
Response: `{"message": "Key revoked."}`

### Developer Generations (Multi-User Tracking)
```
GET /v1/developer/generations?limit=50&offset=0
```
List all generations submitted via YOUR API key. Useful for developers tracking their end-users' usage.

Requires API key auth (not JWT). Returns only generations where `api_key_id` matches caller's key.

Optional query params:
- `limit` (default 50, max 100)
- `offset` (default 0)

Response: `{"generations": [...], "total": 12}`

Each generation includes: id, model, mode, prompt, duration, resolution, aspect_ratio, image_urls, video_urls, token_cost, status, result_url, error_message, created_at, completed_at, created_at_fmt, completed_at_fmt

Note: Only tracks generations made AFTER api_key_id tracking was enabled (2026-07-03).

---

### Kling Elements

Kling Elements are reusable character/object references for consistency across videos.

Create element:
```
POST /v1/kling/elements
```
Body:
- `name` (required): Element name, max 100 chars
- `description` (optional): Description, max 500 chars
- `image_url` (required): Reference image URL
- `tags` (optional): Tags for organization

Cost: 1 token per element.

Response: `{"id": 1, "status": "ready", "element_id": "elem_abc123", "name": "Beg Kuning"}`

List elements:
```
GET /v1/kling/elements
```
Response: `{"elements": [...]}`

Delete element:
```
DELETE /v1/kling/elements/{element_db_id}
```
Response: `{"message": "Element deleted."}`

### Webhooks

Register:
```
POST /v1/webhooks
Body: {"url": "https://yourapp.com/webhook", "secret": "signing_secret", "events": ["video.completed", "image.completed", "job.failed"]}
```
Events: `video.completed`, `image.completed`, `job.failed`, `all` (receive everything).

List:
```
GET /v1/webhooks
```

Delete:
```
DELETE /v1/webhooks/{webhook_id}
```

Webhook payload (video.completed):
```json
{
  "event": "video.completed",
  "task_id": "123",
  "model": "Veo 3.1 Fast",
  "status": "completed",
  "result_url": "https://...",
  "error": null,
  "cost_tokens": 12,
  "timestamp": "2026-07-03T10:30:00Z"
}
```

Webhook payload (job.failed):
```json
{
  "event": "job.failed",
  "task_id": "124",
  "model": "Kling V3.0 Pro",
  "status": "refunded",
  "result_url": null,
  "error": "Model timeout",
  "cost_tokens": 0,
  "timestamp": "2026-07-03T10:31:00Z"
}
```

Signature verification: Header `X-AIVidLab-Signature: sha256=HMAC_SHA256(secret, request_body)`
Failed deliveries are retried 3× with exponential backoff (5s, 30s, 5min).
Max 5 webhooks per account.

---

## AIVIDLAB OMNI — STORYBOARD + VIDEO PIPELINE

Omni is an end-to-end ad creation pipeline: Gemini designs storyboards → GPT Image 2 generates scene sheets → Omni Flash generates videos.

### Omni Styles
```
GET /v1/omni/meta/styles
```
Returns available visual styles.

| Key | Name | Mascot | Description |
|---|---|---|---|
| realistik | Realistik | No | Foto realistik, cinematic |
| ugc | UGC | No | Gaya TikTok UGC, handheld |
| cinematic | Cinematic | No | Cinematic premium, color graded |
| pixar | Pixar 3D | Yes | Gaya Pixar, mascot product |
| kartun3d | Kartun 3D | Yes | Kartun 3D, mascot product |
| anime | Anime | Yes | Gaya anime Jepun |
| claymation | Claymation | Yes | Clay/stop-motion look |

### Create Omni Job
```
POST /v1/omni/create
```
Gemini auto-designs all scenes. Tokens are NOT charged here — only when sheets/videos are generated.

Body:
- `product_name` (optional): Product name (max 200 chars)
- `product_usp` (optional): Unique selling point (max 500 chars)
- `product_cta` (optional, default "Tekan Beg Kuning sekarang"): Call to action (max 100 chars)
- `talent_persona` (optional): Talent description (max 300 chars)
- `mode` (optional, default "product"): "product" or "story"
- `style_key` (optional, default "realistik"): See styles list above
- `scene_count` (optional, default 1, range 1-4): Number of scenes
- `beat_count` (optional, default 3, range 3-4): Beats per scene
- `product_image_url` (optional): Product reference image
- `talent_image_url` (optional): Talent reference image
- `resolution` (optional, default "720p"): "720p" or "1080p"
- `aspect_ratio` (optional, default "9:16"): "9:16" or "16:9"

Response: `{"job_token": "abc123def456", "scenes": [...]}`

### Generate Storyboard Sheets
```
POST /v1/omni/{job_token}/sheets
```
Generates storyboard sheet images (GPT Image 2) for all draft scenes.

Cost: 4 tokens per scene.

Response: `{"submitted": 2, "failed": 0, "message": "2 storyboard dihantar"}`

### Get Omni Job Status
```
GET /v1/omni/{job_token}
```
Returns all scenes with sheet + video status.

### List Omni Jobs
```
GET /v1/omni/jobs
```
Returns all Omni jobs for the current user.

Response: `{"jobs": [...]}`

### Generate Videos
```
POST /v1/omni/{job_token}/videos
```
Generates Omni Flash videos for all scenes with completed sheets.

Cost: 16 tokens per scene.

Response: `{"submitted": 2, "failed": 0, "message": "2 video dihantar"}`

### Regenerate Sheet
```
POST /v1/omni/scene/{scene_id}/regenerate-sheet
```
Regenerate a single scene's storyboard sheet.

Response: `{"task_id": "...", "status": "submitted"}`

### Retry Video
```
POST /v1/omni/scene/{scene_id}/retry-video
```
Retry video generation for a failed scene.

### Edit Scene
```
PATCH /v1/omni/scene/{scene_id}
```
Edit dialog & text-on-screen for a scene (before sheet generation).

Body:
- `vo_dialog` (optional): Full dialog text (max 500 chars)
- `text_on_screen` (optional): Full text-on-screen (max 300 chars)
- `dialog_beats` (optional): Array of per-beat dialog texts (3 or 4 items)
- `text_beats` (optional): Array of per-beat text-on-screen (3 or 4 items)

Response: `{"id": 123, "vo_dialog": "...", "text_on_screen": "...", "dialog_beats": [...], "text_beats": [...]}`

---

## ERROR CODES

| Code | Meaning | Action |
|---|---|---|
| 401 | Invalid/missing API key | Check Bearer token |
| 402 | Insufficient token balance | Top up via /topup |
| 400 | Invalid parameters | Check model ID and params |
| 403 | Account suspended | Contact support |
| 404 | Resource not found | Check task_id or job_token |
| 413 | File too large | Check upload size limits |
| 429 | Rate limited (30 req/min) | Wait and retry |
| 500 | Server error | Retry after 5 seconds |

---

## RATE LIMITS

- 30 requests per minute per API key
- Max 3 API keys per account
- Webhook delivery retries 3× with exponential backoff

---

## WORKFLOW EXAMPLE

1. Submit generation:
```bash
RESPONSE=$(curl -s -X POST https://api.aividlab.shop/v1/video/generate \
-H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model": "grok-video", "prompt": "Your prompt", "duration": 10, "aspect_ratio": "16:9"}')
TASK_ID=$(echo $RESPONSE | jq -r '.task_id')
```

2. Poll until complete:
```bash
while true; do
  STATUS=$(curl -s https://api.aividlab.shop/v1/video/$TASK_ID \
-H "Authorization: Bearer YOUR_API_KEY" \
  echo $STATUS | jq '.status, .progress'
  if echo $STATUS | jq -e '.status == "completed"' > /dev/null; then
    echo "Done! URL:" $(echo $STATUS | jq -r '.result_url')
    break
  fi
  sleep 5
done
```

Or use webhooks to avoid polling entirely.

---

## QUICK REFERENCE — CHEAPEST OPTIONS

- Cheapest video: Grok Video 10s = 7 tokens (RM0.70)
- Cheapest HD video: Kling Standard 3s = 19 tokens (RM1.90)
- Cheapest image: Nano Banana = 2 tokens (RM0.20)
- Best value video: Omni Flash 10s = 16 tokens (RM1.60) with image-to-video support
- Best quality video: Kling 4K Ultra HD 15s = 241 tokens (RM24.10)
- Best quality image: Nano Banana Pro = 6 tokens (RM0.60)
- Cheapest music: Mureka 7.6 Song = 5 tokens (RM0.50)
- Best music: Suno V5.5 = 9 tokens (RM0.90)
- Cheapest lyrics: Suno Lyrics / Mureka Lyrics = 4 tokens (RM0.40)
- Cheapest BGM: Mureka 7.6 BGM = 5 tokens (RM0.50)
- Cheapest TTS: Speech-02 Turbo = 5 tokens (RM0.50)
- Best TTS: Speech-2.8 HD = 6 tokens (RM0.60)

---

## VIDEO GENERATE — FULL PARAMETER REFERENCE

| Parameter | Type | Required | Description |
|---|---|---|---|
| model | string | yes | Model ID (see models list) |
| prompt | string | yes | Text description (max 5000) |
| duration | int | no | Seconds (default 10, range 3-30) |
| resolution | string | no | "480p", "720p", "1080p", "2k", "4k", "native1080p", "native4k" |
| aspect_ratio | string | no | "16:9", "9:16", "1:1" (default "9:16") |
| image_url | string | no | Reference image URL (i2v mode) |
| video_url | string | no | Reference video URL (multimodal/ref2v/video_edit/motion) |
| mode | string | no | Kling: "t2v", "i2v", "ref2v", "video_edit", "motion" |
| image_urls | array | no | Multiple ref images (max 9, multimodal/ref2v) |
| last_image_url | string | no | End frame image (Kling i2v) |
| orientation | string | no | Motion: "image" or "video" |
| sound | bool | no | Enable sound sync (Kling Pro auto) |
| negative_prompt | string | no | What to avoid (Kling, max 5000) |
| cfg_scale | float | no | CFG scale (Kling) |
| shot_type | string | no | Shot type (Kling, default "customize") |
| element_list | array | no | Kling element IDs (max 3) |