> ## Documentation Index
> Fetch the complete documentation index at: https://developers.fastdrop.io/llms.txt
> Use this file to discover all available pages before exploring further.

# What you can ask for, and what it costs

> The check that decides whether to spend, and the capabilities worth spending on

<Note>
  **Fixing is priced differently.** Every other row on this page costs a flat
  number of credits per video. [Fixing a video](/api-reference/fix-video) is billed
  per minute of output by resolution, because that is what it costs us — see
  [Pricing](/pricing) for the rates. It carries no flat number below for that
  reason, not because it is unavailable.
</Note>

FastDrop uses a credit-based pricing system. Every API call consumes credits based on which capabilities you request and your processing tier.

## Before you commit

One call, and it decides whether any of the rest is worth making:

| Capability | Credits | Description |
| - | - | - |
| `readiness` | 1 | Decide whether footage suits an intended use. Returns a verdict — `proceed`, `fix_first` or `reject` — with machine-readable fix instructions, inline in the same request rather than as a job. See [Why check footage before you process it](/guides/checking-footage) |
| [`fix`](/api-reference/fix-video) | by duration | Execute the fix that verdict prescribed, then re-run the check on the result. Failed fixes cost nothing |

`readiness` is deliberately the cheapest call in the API, because it is the one
that stops you making the expensive ones. It is not a quality score: the same
file is fit for one job and unfit for another, so what you intend to do with it
is part of the question.

## Once footage passes

Each of these analyzes a different aspect of your video, and none of them
inspect whether the video was suitable in the first place:

| Capability | Credits | Description |
| - | - | - |
| `classify` | 2 | Assign one of 6 editor roles with confidence score, explanation, suggested filename, and folder path |
| `thumbnails` | 1 | Extract 3-5 keyframe thumbnails from the video |
| `diagnostics` | 1 | Get codec, resolution, FPS, VFR detection, duration, health score, and container format |
| `duplicates` | 1 | Generate a perceptual hash fingerprint for duplicate detection across your library |
| `clips` | 2 | Detect clip candidate timestamps with labels, explanations, and confidence scores |
| `transcribe` | 8 | Speech transcription (50+ languages), English translation, optional SRT/TXT file generation |

<Note>
  `diagnostics` and `readiness` are easy to confuse and answer different
  questions. Diagnostics reports what the file *is* — codec, resolution, frame
  rate. Readiness reports what you should *do* about it, given a purpose. If you
  want a decision, ask for readiness; if you want the raw measurements behind
  one, ask for diagnostics.
</Note>

### Classification roles

`classify` assigns one of six editor-ready roles, each with a confidence score, explanation, suggested filename and folder path:

| Role | Description |
| - | - |
| **Hook** | Attention-grabbing opening clips |
| **A-Roll** | Primary speaker or interview footage |
| **B-Roll** | Supplementary cutaway footage |
| **Screen Recording** | Screen captures and tutorials |
| **VO** | Voice-over audio clips |
| **Blooper** | Outtakes and mistakes |

### Search (post-processing)

After transcription, you can search across your videos at no additional cost:

| Action | Credits | Description |
| - | :-: | - |
| Batch transcribe (for search) | 3/video | Transcribe videos during batch processing for search indexing |
| Transcript search query | 0 | Search across transcribed segments by keyword or meaning |
| Visual search query | 0 | Search by visual description using CLIP similarity |

See the [Transcript Search](/guides/search) and [Visual Search](/guides/visual-search) guides for details.

You can request any combination of capabilities per video. Credits are summed:

```json theme={null}
// classify (2) + thumbnails (1) + diagnostics (1) = 4 credits
{
  "capabilities": ["classify", "thumbnails", "diagnostics"]
}
```

## Processing tiers

The processing tier controls how classification is performed. It only affects the `classify` capability — other capabilities always run the same way.

| Tier | Credit cost | Method |
| - | - | - |
| `basic` | 2 credits | Heuristic analysis only. Fastest, uses motion detection, face detection, and audio analysis. |
| `enhanced` | 4 credits | Heuristic first, falls back to AI if confidence is low. Best balance of speed and accuracy. |
| `premium` | 6 credits | Full AI analysis via GPT-4 Vision. Highest accuracy, slower. |

<Note>
  The credit cost listed above is for the `classify` capability only. Other capabilities (thumbnails, diagnostics, etc.) cost the same regardless of tier.
</Note>

## Bundles

Bundles group common capabilities at a discount:

| Bundle | Credits | Includes |
| - | - | - |
| `full_pipeline` | 6 | `classify` (enhanced), `thumbnails`, `diagnostics` |

Use a bundle by passing its name in the capabilities array:

```json theme={null}
{
  "capabilities": ["full_pipeline"]
}
```

## Plans

| Plan | Monthly price | Credits/month | Rate limit | Max concurrent | Max file size |
| - | :-: | :-: | :-: | :-: | :-: |
| **Free** | \$0 | 100 | 10 req/min | 3 | 500 MB |
| **Starter** | \$9 | 150 | 20 req/min | 5 | 1 GB |
| **Growth** | \$49 | 1,000 | 40 req/min | 15 | 2 GB |
| **Scale** | \$179 | 5,000 | 80 req/min | 50 | 5 GB |

Credits reset monthly. Unused credits do not roll over.

## Credit examples

| Workflow | Capabilities | Tier | Credits/video |
| - | - | - | :-: |
| Quick classify only | `classify` | basic | 2 |
| Classify + thumbnails | `classify`, `thumbnails` | basic | 3 |
| Standard workflow | `classify`, `thumbnails`, `diagnostics` | basic | 4 |
| Full pipeline bundle | `full_pipeline` | enhanced | 6 |
| Everything | `classify`, `thumbnails`, `diagnostics`, `duplicates`, `clips` | basic | 7 |
| Everything + transcribe | `classify`, `thumbnails`, `diagnostics`, `duplicates`, `clips`, `transcribe` | basic | 15 |
| Everything at premium | `classify`, `thumbnails`, `diagnostics`, `duplicates`, `clips`, `transcribe` | premium | 19 |

A `readiness` check costs 1 credit and answers before you commit to any of the above.

## Checking your usage

Call the [usage endpoint](/api-reference/check-usage) to see your current balance:

```bash theme={null}
curl https://api.fastdrop.io/api/v1/usage \
  -H "X-API-Key: fd_live_your_key_here"
```

```json Response theme={null}
{
  "plan_tier": "starter",
  "credits_remaining": 126,
  "credits_monthly": 150,
  "credits_reset_at": "2026-04-01T00:00:00Z",
  "usage_this_month": {
    "total_credits": 24,
    "total_requests": 8,
    "by_capability": {
      "classify": 16,
      "thumbnails": 4,
      "diagnostics": 4,
      "duplicates": 0,
      "clips": 0,
      "transcribe": 0
    }
  }
}
```

## Listing capabilities programmatically

The [capabilities endpoint](/api-reference/list-capabilities) returns current costs and bundles. No API key required:

```bash theme={null}
curl https://api.fastdrop.io/api/v1/capabilities
```


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.