Pexels Synthetic Captions — InternVL2-40B
Source, license and coverage
Supplier documentation. These claims are separate from the automated sample score. A listing edit date is not a data freshness date.
- License
- cc-by-sa-4.0
- Source / creator
- CaptionEmporium/pexels-568k-internvl2
- Collection method
- The creator took the photo-concept-bucket image set (Pexels images with associated tags), passed each image through InternVL2-40B-AWQ for caption generation, and used the original tags as grounding context to reduce hallucination and improve specificity. No manual review of captions is documented at scale.
- Coverage start
- Not documented
- Coverage end
- Not documented
- Data last updated
- Not documented
- Update schedule
- Not documented
Captions are synthetic and inherit any biases, hallucinations, or stylistic quirks of InternVL2-40B-AWQ. Primarily English, but occasionally transcribes in-image text in other languages. Pexels content skews toward stock-photo aesthetics (clean compositions, common subjects) and is not representative of general web imagery. The dataset is not deduplicated against common text-to-image evaluation suites. Source does not document per-caption quality scores; buyers should validate empirically before fine-tuning.
Sample structure score: 96.8 / 100
This automated check describes the inspected sample, not factual accuracy, legal rights, representativeness, or the quality of the entire dataset. It is not a customer rating.
Assessed 10 sample records (JSON) on 2026-10-09. All records in the provided sample were checked.
| Check | Points | Evidence |
|---|---|---|
| Populated cells | 46.8 / 50 | 187 of 200 top-level cells contain a value. Null, absent and blank values count as missing; zero and false count as populated. |
| Consistent value types | 30 / 30 | 187 of 187 populated cells match their column's most common observed type. Types are inferred, not checked against real-world truth. |
| Consistent record shape | 20 / 20 | 10 of 10 records have the expected fields. CSV/TSV use the header width; JSON uses the union of observed keys. |
Field-level findings and improvements
Check missing cells and mixed types below. Document intentional missing values or mixed types in your field descriptions. Do not fill legitimate unknowns with invented values just to increase this score.
| Field | Missing cells | Most common type | Other populated types |
|---|---|---|---|
| id | 0 / 10 | number | 0 / 10 |
| class_label | 0 / 10 | string | 0 / 10 |
| type | 0 / 10 | string | 0 / 10 |
| slug | 0 / 10 | string | 0 / 10 |
| description | 10 / 10 | unknown | 0 / 0 |
| alt | 0 / 10 | string | 0 / 10 |
| created_at | 0 / 10 | string | 0 / 10 |
| title | 0 / 10 | string | 0 / 10 |
| location | 3 / 10 | string | 0 / 7 |
| tags | 0 / 10 | string | 0 / 10 |
| main_color | 0 / 10 | object | 0 / 10 |
| colors | 0 / 10 | object | 0 / 10 |
| width | 0 / 10 | number | 0 / 10 |
| height | 0 / 10 | number | 0 / 10 |
| aspect_ratio | 0 / 10 | number | 0 / 10 |
| url | 0 / 10 | string | 0 / 10 |
| cogvlm_caption | 0 / 10 | string | 0 / 10 |
| megapixels | 0 / 10 | number | 0 / 10 |
| __index_level_0__ | 0 / 10 | number | 0 / 10 |
| internvl2_caption | 0 / 10 | string | 0 / 10 |
About this data
English captions for Pexels photos generated with InternVL2-40B-AWQ and grounded with original tags. Suitable for text-to-image and image-to-text model training.
Retrieve with your agent or Python
Create an account and configure DATABAZAAR_API_KEY. This example retrieves free or already purchased data; it never makes a purchase. For a multi-file dataset, choose a file index from the manifest.
Download the Python examplepython3 retrieve-dataset.py 1c86af8b-445b-4a7b-8358-d3f9bbeb9ad4 --output dataset.bin
Full supplier documentation
Schema
| Name | Type | Description |
|---|---|---|
| id | BIGINT | |
| class_label | VARCHAR | |
| type | VARCHAR | |
| slug | VARCHAR | |
| description | VARCHAR | |
| alt | VARCHAR | |
| created_at | VARCHAR | |
| title | VARCHAR | |
| location | VARCHAR | |
| tags | VARCHAR | list[string] — original Pexels tags used to ground the caption |
| main_color | BIGINT[] | |
| colors | VARCHAR[] | |
| width | BIGINT | |
| height | BIGINT | |
| aspect_ratio | DOUBLE | |
| url | VARCHAR | |
| cogvlm_caption | VARCHAR | |
| megapixels | DOUBLE | |
| __index_level_0__ | BIGINT | |
| internvl2_caption | VARCHAR |
Sample Data
Preview a sample of the data before downloading.
Public sample only. Sign in to retrieve the full dataset, including free datasets.
For AI Agents
# 1. Add to your agent's MCP config (claude_desktop_config.json or similar):
{
"mcpServers": {
"databazaar": { "command": "npx", "args": ["databazaar-mcp"] }
}
}
# 2. Your agent can then call:
search_datasets({ query: "Pexels Synthetic Captions — In" })
// Found: 1c86af8b-445b-4a7b-8358-d3f9bbeb9ad4
get_download_url({ dataset_id: "1c86af8b-445b-4a7b-8358-d3f9bbeb9ad4" }) // free — sign in with MCP OAuth first# Free dataset — sign in or use your account API key: curl https://api.databazaar.io/datasets/1c86af8b-445b-4a7b-8358-d3f9bbeb9ad4/download-url -H "Authorization: Bearer $DATABAZAAR_API_KEY"