Model reference · open weights
Vela-1.0-Encoder-Embedding is an open-weight embedding model from llm-semantic-router. AxForge deploys and operates it for you on dedicated EU-owned hardware — with the licence handled where one is required.
Available as managed deployment — configured and operated for you on dedicated EU hardware, quoted per deployment.
What it is
| Released by | llm-semantic-router |
|---|---|
| Type | Embedding models |
| Task | Embeddings |
| Parameters (lead) | 307M |
| Context | 32k tokens |
| Runs with | sentence-transformers |
| Based on | llm-semantic-router/Vela-1.0-Encoder-307M |
| Released | 2026-09-12 |
| Popularity | 672 downloads / month |
| Licence | Open weights |
About
Connect requests with the right context. Vela Embedding powers multilingual semantic matching, retrieval, and memory in your router.
307M parameters · 32K context · Multilingual
Compared with the original mmBERT Embedding. Scores are ×100; higher is better.
Selected tasks from MTEB 2.21.0:
| Task | Original mmBERT | Vela |
|---|---|---|
| Banking77 | 72.26 | 73.76 |
| MASSIVE · English | 63.51 | 68.24 |
| MASSIVE · Chinese | 61.41 | 66.06 |
| DuRetrieval | 62.47 | 62.18 |
| NFCorpus | 26.38 | 26.11 |
| SciFact | 46.81 | 47.64 |
| PAWS-X · English | 62.98 | 71.80 |
| PAWS-X · Chinese | 59.29 | 62.91 |
| STS-B · Chinese | 75.00 | 74.61 |
| STS-B · English | 80.38 | 81.06 |
| CLS Clustering | 34.20 | 34.55 |
| Twenty Newsgroups Clustering | 38.05 | 39.79 |
| Tatoeba · Chinese–English | 89.13 | 88.72 |
Long-document retrieval:
| Evaluation | Metric | Original mmBERT | Vela |
|---|---|---|---|
| SummScreenFD · 336 queries | nDCG@10 | 76.99 | 88.47 |
Both models use 22 layers, 768 dimensions and FP32. The MTEB rows use each task's standard main metric; SummScreenFD uses its validation split.
These results describe the listed tasks, not a full MTEB score or ranking.
Install sentence-transformers. Use a ROCm-enabled PyTorch build for AMD GPUs.
import torch
from sentence_transformers import SentenceTransformer
model_id = "llm-semantic-router/Vela-1.0-Encoder-307M-Embedding"
device = "cuda" if torch.cuda.is_available() else "cpu"
model = SentenceTransformer(model_id, device=device)
vectors = model.encode([
"The library opens in the morning.",
"图书馆早上开门。",
])
print(vectors.shape) # (2, 768)
The default returns 768-dimensional unit vectors. The 32,768-token input limit includes special tokens.
From the published model card. Full card on the HuggingFace links in the sidebar.
Using it via the API
Once AxForge deploys vela-1-0-encoder-embedding for you, it answers on the OpenAI-compatible API — the same base URL and keys as every other model. (vela-1-0-encoder-embedding below is illustrative; you get the exact model name on deployment.)
$ curl -sS https://api.axforge.ai/v1/embeddings \
-H "Authorization: Bearer $AXFORGE_API_KEY" \
-H "Content-Type: application/json" \
-d '{"model":"vela-1-0-encoder-embedding","input":"text to embed"}'
Create an account — your API key is available in the console. 3M free tokens every 30 days with every new account.