Decision-2.0-Eos-0.8B
Decision-2.0-Eos-0.8B is the 0.8B model of Decision 2.0, the decision models of vLLM Semantic Router. Give it an input (text or JSON) and the questions you need answered: pick one of several options, say yes or no, or rate on a scale. It answers them all at once and returns a probability for every answer, without generating text.
| Parameters | 0.75B |
| Context length | 16,384 tokens |
| Decision types | Choice · Yes / No · Score |
| License | Apache-2.0 |
Highlights
- Top JevArena score of its size: 53.9, ahead of the 3 other same-size models compared.
- Ahead of Decision 1.0 Eos: +11.4 on JevArena and +1.7 on the Jev Decision Index.
- Speed: a median of 6.0 ms per single-question request on a single GPU.
- Many questions, one pass: Choice, Yes / No and Score questions about the same input are answered together in one forward pass, with a probability for every option.
Quickstart
pip install "transformers>=5.17" torch safetensors
import json
from transformers import AutoModel
model = AutoModel.from_pretrained("vllm-sr/Decision-2.0-Eos-0.8B", trust_remote_code=True)
result = model.system_one(
state="The order arrived damaged yesterday. The customer has a receipt and asks for a replacement today.",
questions={
"route": {
"type": "choice",
"instructions": "Which team should handle this request?",
"criteria": {
"returns": "Refunds, replacements and damaged deliveries",
"billing": "Payments, invoices and charges",
"technical": "Product setup and faults"
}
},
"receipt": {
"type": "noul",
"instructions": "Does the customer have a receipt?"
},
"urgency": {
"type": "score",
"instructions": "How urgent is this request?",
"criteria": [
"Routine",
"Soon",
"Today"
]
}
},
)
print(json.dumps(result["answers"], indent=2))
# Or as a pipeline:
# transformers.pipeline("decision", model="vllm-sr/Decision-2.0-Eos-0.8B", trust_remote_code=True)(state=..., questions=...)
Evaluation
| Model | JevArena ↑ | Human-labelled transfer ↑ | Jev Decision Index ↑ |
|---|---|---|---|
| Decision-2.0-Eos-0.8B | 53.9 | 50.3 | 20.1 |
| Intern-Decision-0.8B | 43.5 | 38.2 | — |
| Kev-0.8B | 43.2 | 39.0 | — |
| Decision 1.0 Eos | 42.5 | 46.1 | 18.4 |
JevArena
Every model answers the same frozen prompts, scored the same way; missing or invalid answers count as errors. Human-labelled transfer is the median macro-F1 over 15 human-labelled tasks (×100).
Jev Decision Index
Decision 2.0: independent reproduction with the official 0.2.1 kit on the released weights; others: public board snapshot, 2026-09-28. Training data audited at row level against all Index test items.
License
Apache-2.0 (LICENSE).
Citation
@misc{decision_2_0_eos_0_8b_2026,
title = {{Decision-2.0-Eos-0.8B}: A Decision 2.0 Model for Structured Decisions},
author = {{vLLM Semantic Router Team}},
year = {2026},
howpublished = {\url{https://huggingface.co/vllm-sr/Decision-2.0-Eos-0.8B}}
}