Visual Document Retrieval
Safetensors
ColPali
sentence-transformers
colpali_engine
qwen3_5
multimodal-retrieval
late-interaction
colqwen
ColQwen3_5
vidore
mteb
qwen3.5
model-merge
per-block-merge
MaxSim
multi-vector
Instructions to use vultr/VultronRetrieverPrime-Qwen3.5-8B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- ColPali
How to use vultr/VultronRetrieverPrime-Qwen3.5-8B with ColPali:
# No code snippets available yet for this library. # To use this model, check the repository files and the library's documentation. # Want to help? PRs adding snippets are welcome at: # https://github.com/huggingface/huggingface.js
- sentence-transformers
How to use vultr/VultronRetrieverPrime-Qwen3.5-8B with sentence-transformers:
from sentence_transformers import SentenceTransformer model = SentenceTransformer("vultr/VultronRetrieverPrime-Qwen3.5-8B") sentences = [ "The weather is lovely today.", "It's so sunny outside!", "He drove to the stadium." ] embeddings = model.encode(sentences) similarities = model.similarity(embeddings, embeddings) print(similarities.shape) # [3, 3] - Notebooks
- Google Colab
- Kaggle
Update README.md
Browse files
README.md
CHANGED
|
@@ -42,7 +42,7 @@ footprint-sensitive serving. Trained and evaluated on [Vultr Cloud](https://www.
|
|
| 42 |
|
| 43 |
## Highlights
|
| 44 |
|
| 45 |
-
- **#1 on ViDoRe V3:** 64.26 mean nDCG@10 over all 10 tasks (8 public + 2 private), at
|
| 46 |
- **Family 1–2 finish:** ranks 1 and 2 on V3 are both VultronRetriever; [Core](https://huggingface.co/vultr/VultronRetrieverCore-Qwen3.5-4.5B) holds #2 at 63.57.
|
| 47 |
- **#1 on six of the ten V3 tasks** (Energy, FinanceEn, FinanceFr, Industrial, Pharmaceuticals, Physics).
|
| 48 |
- **Official MTEB:** V1 92.08, V2 68.18, V3 64.26.
|
|
|
|
| 42 |
|
| 43 |
## Highlights
|
| 44 |
|
| 45 |
+
- **#1 on ViDoRe V3:** 64.26 mean nDCG@10 over all 10 tasks (8 public + 2 private), at up to 16× smaller index than the 2560–4096-dim field.
|
| 46 |
- **Family 1–2 finish:** ranks 1 and 2 on V3 are both VultronRetriever; [Core](https://huggingface.co/vultr/VultronRetrieverCore-Qwen3.5-4.5B) holds #2 at 63.57.
|
| 47 |
- **#1 on six of the ten V3 tasks** (Energy, FinanceEn, FinanceFr, Industrial, Pharmaceuticals, Physics).
|
| 48 |
- **Official MTEB:** V1 92.08, V2 68.18, V3 64.26.
|