# VTXAI/vtx-embed-7M

Model from Hugging Face for feature extraction: VTXAI/vtx-embed-7M · tokenizers

## Facts

- Type: Embedding
- Task: feature-extraction
- Framework: tokenizers
- License: mit — Permissive open license: commercial use, modification, and redistribution are permitted. (https://spdx.org/licenses/MIT.html)
- Source: https://huggingface.co/VTXAI/vtx-embed-7M
- Hosting: hosted (3 files)
- Files: 3 hosted of 9 listed
- File counts: "listed" is every file in the source repository; only hosted files download from Hugging Bay, and the download plan lists the files it selects for one download, so its count can differ.
- Hosted here: 3 files, 5.41 MB
- Full repo upstream: 5.53 MB
- Recommended file: 4.72 MB · model.safetensors
- Parameters: 7M
- Download: https://huggingbay.xyz/artifact/hf-model-vtxai-vtx-embed-7m/files (download plan: https://huggingbay.xyz/api/artifacts/hf-model-vtxai-vtx-embed-7m/download-plan)
- Per-file SHA-256: available for 3 of 3 hosted files via https://huggingbay.xyz/api/artifacts/hf-model-vtxai-vtx-embed-7m/files?limit=100
- Run-ready: no (Hugging Bay has not certified a run path for this model; hosted bytes are downloadable and hash-checkable only)
- Peer-assisted fallbacks: 0 verified, 0 pending review
- Downloads: 87
- Stars: 2
- Card completeness: 96%

## Description

External Hugging Face catalog record for VTXAI/vtx-embed-7M. Task: feature extraction. Library: tokenizers. Languages: en. Includes 6 model-card evaluation signals.

## Model Card Facts

- Declared license: mit
- Pipeline: feature-extraction
- Library: tokenizers
- Task categories: feature-extraction
- Languages: en
- Metrics: spearman_cosine
- Model-index entries: vtx-embed-7M
- Evaluation signals: 6 model-card eval results
- File hints: 9 upstream file samples

## Evaluation Signals

- STSBenchmark: cosine_spearman = 0.7918 (Semantic Textual Similarity)
- SICK-R: cosine_spearman = 0.6294 (Semantic Textual Similarity)
- Banking77Classification: accuracy = 0.7043 (Classification)
- AmazonCounterfactualClassification: accuracy = 0.6679 (Classification)
- TwentyNewsgroupsClustering: v_measure = 0.2936 (Clustering)
- RedditClustering: v_measure = 0.388 (Clustering)

## File Hints

- .gitattributes (1.5 KB)
- README.md (5.4 KB)
- __pycache__/vortex_embed_v4_5.cpython-312.pyc (16 KB)
- benchmark.py (26 KB)
- benchmark_results.json (58 KB)
- config.json (251 B)
- model.safetensors (4.5 MB) LFS
- tokenizer.json (668 KB)
- vortex_embed_v4_5.py (11 KB)

## Quality Checklist

- Summary: passed
- Description: passed
- Model card metadata: passed
- License: passed
- Provenance: passed
- File inventory: passed
- Hosted bytes: passed
- Scan result: passed
- Manifest signature: passed

## Safety Notes

- External catalog metadata imported from Hugging Face public API on 2026-09-29T22:38:40.342Z.
- Open license detected as mit.
- Model-index metadata found for vtx-embed-7M.
- Not a run kit: hosted, hash-checked bytes only; run readiness is not certified for this model yet.

## Actions

- Web: https://huggingbay.xyz/artifact/hf-model-vtxai-vtx-embed-7m
- Files API: https://huggingbay.xyz/api/artifacts/hf-model-vtxai-vtx-embed-7m/files?limit=100
- Metadata: https://huggingbay.xyz/api/artifacts/hf-model-vtxai-vtx-embed-7m/metadata.json
- Distribution: https://huggingbay.xyz/api/artifacts/hf-model-vtxai-vtx-embed-7m/distribution
- Manifest: https://huggingbay.xyz/api/manifests/hf-model-vtxai-vtx-embed-7m
- Download plan (per-file SHA-256 + signed download URLs): https://huggingbay.xyz/api/artifacts/hf-model-vtxai-vtx-embed-7m/download-plan
- Hosted download rows: https://huggingbay.xyz/api/downloadable-files?artifactId=hf-model-vtxai-vtx-embed-7m&limit=100
- Request mirror: not applicable.
- Submit peer fallback for review: POST https://huggingbay.xyz/api/artifacts/hf-model-vtxai-vtx-embed-7m/decentralized-fallbacks