Zero-Shot Classification
Transformers
Safetensors
qwen3_5
feature-extraction
decision-model
classification
system-one
multimodal
vision
video
custom_code
Instructions to use vllm-sr/d3.1-nano with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use vllm-sr/d3.1-nano with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("zero-shot-classification", model="vllm-sr/d3.1-nano", trust_remote_code=True)# pip install -U transformers accelerate # Load model directly from transformers import AutoProcessor, AutoModel processor = AutoProcessor.from_pretrained("vllm-sr/d3.1-nano", trust_remote_code=True) model = AutoModel.from_pretrained("vllm-sr/d3.1-nano", trust_remote_code=True, device_map="auto") - Notebooks
- Google Colab
- Kaggle
Add missing identity.model_sha256 to MODEL_MANIFEST.json
#1
by ingvarson - opened
Hi! vllm-srun serve refuses to load this model:
loading d3.1-nano failed: model identity differs from the scored checkpoint in MODEL_MANIFEST.json
The manifest has no identity.model_sha256 field. The other d3 packages (d3-nano, d3-turbo) have it, and the runtime checks it on load.
This PR adds that one line. I got the value from the runtime's own model_identity(), using the file hashes and the decision config that are already in the manifest. For d3-nano the same function gives exactly the value from its manifest, so it should be right.
With this change the model loads and answers fine for me (semantic-router main, e0008e2).
Thanks! Verified: model_identity() at semantic-router e0008e2 gives the same value for this package. Merging.
TheoHsiung changed pull request status to merged