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

# Pricing

> Per-request pricing with no token counting. Every response includes its cost.

## TL;DR pricing

| What                   | Cost                |
| ---------------------- | ------------------- |
| Chat (budget models)   | \$0.003/request     |
| Chat (standard models) | \$0.006/request     |
| Chat (premium models)  | \$0.015/request     |
| Chat (ultra models)    | \$0.03/request      |
| Images                 | $0.03 – $0.10/image |
| Video                  | $3.00 – $5.00/video |
| Search                 | \$0.05/query        |

No token counting. No complicated tiers. Every response includes the exact cost:

```json theme={null}
{ "cost": { "this_request": "$0.006" } }
```

## Credits model

Buy credits upfront. 1 credit = \$0.01. Credits never expire.

| Pack           | Price | What you get                                           |
| -------------- | ----- | ------------------------------------------------------ |
| 1,000 credits  | \$10  | \~1,666 budget chats, or \~333 images, or 3 videos     |
| 2,500 credits  | \$25  | \~4,166 budget chats, or \~833 images, or 8 videos     |
| 10,000 credits | \$100 | \~16,666 budget chats, or \~3,333 images, or 33 videos |

[Add credits →](https://www.ninjachat.ai/developers/billing)

### Example: building a chatbot

A chatbot handling **100 conversations/day** with GPT-5 (\$0.006/request):

* Daily cost: 100 × $0.006 = **$0.60/day\*\*
* Monthly cost: \~**\$18/month**
* Recommended pack: **\$25** (covers \~6 weeks)

A high-volume classifier using DeepSeek V3 (\$0.003/request):

* 10,000 requests/day: **\$30/day**
* Recommended pack: **\$100** (covers \~3 days)

***

## All model costs

### Chat — `POST /api/v1/chat`

**\$0.03/req (ultra):** `gpt-5.4-pro` `claude-opus-4.6`

**\$0.015/req (premium):** `gpt-5.4` `claude-sonnet-4.6`

**\$0.006/req (standard):** `gpt-5` `o3-mini` `claude-sonnet-4.5` `claude-haiku-4.5` `gemini-2.5-pro` `gemini-3-pro` `gemini-3.1-pro` `grok-4` `kimi-k2` `mistral-large` `llama-4-maverick` `kimi-k2.5`

**\$0.003/req (budget):** `gpt-5-mini` `gemini-2.5-flash` `gemini-3-flash` `llama-4-scout` `deepseek-v3` `qwq-32b` `glm-5` `minimax-m2.5` `ninja-1` `uncensored-ai`

### Images — `POST /api/v1/images`

| Model              | Cost       |
| ------------------ | ---------- |
| `flux-kontext-max` | \$0.10/img |
| `flux-2-flex`      | \$0.08/img |
| `flux-1-pro-ultra` | \$0.08/img |
| `recraft-v3`       | \$0.08/img |
| `google-imagen-4`  | \$0.08/img |
| `nano-banana-pro`  | \$0.08/img |
| `seedream`         | \$0.08/img |
| `flux-2-pro`       | \$0.05/img |
| `flux-kontext-pro` | \$0.05/img |
| `nano-banana-2`    | \$0.05/img |
| `flux-1-fill`      | \$0.05/img |
| `flux-2-klein`     | \$0.03/img |
| `nano-banana`      | \$0.03/img |

Generating `n` images = `n` × per-image cost.

### Video — `POST /api/v1/video`

| Model                                                                        | Cost         |
| ---------------------------------------------------------------------------- | ------------ |
| `veo-3.1`                                                                    | \$5.00/video |
| `seedance-2` `kling-video` `google-veo-2` `veo-3.1-fast` `google-veo-3-fast` | \$3.00/video |

Failed generations are refunded automatically.

### Search — `POST /api/v1/search`

**\$0.05/query.**

***

## When credits run out

You get a `402` error:

```json theme={null}
{ "error": "insufficient_credits", "message": "Insufficient credits." }
```

No request goes through. [Add more credits](https://www.ninjachat.ai/developers/billing) and try again.

## Check your balance

From code:

```python theme={null}
r = requests.get("https://www.ninjachat.ai/api/api-keys",
    headers={"Authorization": f"Bearer {os.environ['NINJACHAT_API_KEY']}"})
print(r.json()["billing"]["balanceFormatted"])  # → "$42.50"
```

Or from your [Developers dashboard](https://www.ninjachat.ai/developers).

## Low-balance warnings

When your balance drops below \$5.00, every API response includes a `balance_warning` field:

```json theme={null}
{
  "message": { "role": "assistant", "content": "..." },
  "cost": { "this_request": "$0.006" },
  "balance": "$3.50",
  "balance_warning": {
    "warning": "low_balance",
    "threshold": "Balance is $3.50. Consider adding credits soon."
  }
}
```

At \$1.00 or less, the warning becomes `"critical_low_balance"`. Use this to trigger alerts in your app before hitting `insufficient_credits`.

## Purchase history

View all past credit purchases:

```python theme={null}
r = requests.get("https://www.ninjachat.ai/api/api-keys/transactions",
    headers={"Authorization": f"Bearer {os.environ['NINJACHAT_API_KEY']}"})

for t in r.json()["transactions"]:
    print(f"{t['timestamp']} — {t['amount']} ({t['type']})")
```

Supports `limit` and `offset` query params for pagination.

## Usage logs

Get per-request usage logs with filtering:

```python theme={null}
# Recent logs
r = requests.get("https://www.ninjachat.ai/api/api-keys/logs",
    headers={"Authorization": f"Bearer {os.environ['NINJACHAT_API_KEY']}"},
    params={"limit": 50, "group": "chat"})

for log in r.json()["logs"]:
    print(f"{log['timestamp']} — {log['model']} — {log['statusCode']} — {log['latencyMs']}ms")
```

### Filter options

| Param    | Description                | Example                             |
| -------- | -------------------------- | ----------------------------------- |
| `group`  | Endpoint type              | `chat`, `search`, `images`, `video` |
| `model`  | Model ID                   | `gpt-5`, `flux-2-klein`             |
| `status` | HTTP status code           | `200`, `402`                        |
| `from`   | Start date (ISO)           | `2026-03-01T00:00:00Z`              |
| `to`     | End date (ISO)             | `2026-03-06T23:59:59Z`              |
| `limit`  | Results per page (max 500) | `100`                               |
| `offset` | Pagination offset          | `100`                               |

### CSV export

Add `format=csv` to download as a spreadsheet:

```bash theme={null}
curl "https://www.ninjachat.ai/api/api-keys/logs?format=csv&from=2026-03-01" \
  -H "Authorization: Bearer nj_sk_YOUR_API_KEY" \
  -o usage.csv
```

For rate limits per endpoint, see [Rate Limits](/rate-limits).