Jev directory
CommunityJev Models & Reimplementations
Decision-2.0-Kai-0.6B
The 0.6B model in vLLM Semantic Router's Decision 2.0 family: a 0.60B fine-tune of Qwen3-0.6B-Base with an 8,192-token context that returns probabilities for choice, yes/no and score questions from one forward pass.
- Reading time
- 2 min
- Published by
- Community
- Author
- vLLM Semantic Router
- Last verified
- Oct 2026
- Capabilities
- Open sourceSelf-hostedRuns locally
Highlights
- 0.60B parameters with an 8,192-token context, Apache-2.0, fine-tuned from Qwen/Qwen3-0.6B-Base.
- Card reports JevArena 48.6, the top score of its size class among five models compared.
- 12.7 JevArena and 9.8 Decision Index ahead of Decision 1.0 Kai; human-labelled transfer 45.9.
- Median 4.9 ms per single-question request on a single GPU, all question types in one forward pass.
- Quickstart: transformers>=5.17, torch, safetensors, then AutoModel...system_one or a 'decision' pipeline.
Quickstart
python
import json
from transformers import AutoModel
model = AutoModel.from_pretrained("vllm-sr/Decision-2.0-Kai-0.6B", 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"}},
"receipt": {"type": "noul", "instructions": "Does the customer have a receipt?"}},
)
print(json.dumps(result["answers"], indent=2))Watch out
Card reviewed 2026-10-17; every score and the 4.9 ms median are self-reported, with competitor rows drawn from a public board snapshot dated 2026-09-28, and the card does not name the GPU or the request mix behind the timing.
Questions about this entry
Short answers drawn from the linked project or page.
The 0.6B model of Decision 2.0, vLLM Semantic Router's decision-model family, fine-tuned from Qwen/Qwen3-0.6B-Base. It answers choice, yes/no and score questions about an input at once and returns a probability for every answer instead of text.
It takes typed questions about a state and returns probabilities in one forward pass without output tokens, matching the System One shape. The quickstart calls model.system_one(state=..., questions=...) and prints the answers dictionary.
JevArena 48.6, the top score of its size class among five models compared, plus human-labelled transfer 45.9 and Jev Decision Index 16.3, up 9.8 from Decision 1.0 Kai's 6.5. All are self-reported.
A median of 4.9 ms per single-question request on a single GPU, with every question of the request sharing one forward pass. The card gives no GPU model or batching detail behind the median.
0.60B parameters and an 8,192-token context; Apache-2.0, fine-tuned from Qwen/Qwen3-0.6B-Base, carrying the zero-shot-classification pipeline tag.
Install transformers 5.17 or newer with torch and safetensors, then load it with AutoModel.from_pretrained(..., trust_remote_code=True) and call system_one; the card also shows a transformers pipeline('decision', ...) alternative.
The table lists GLiNER2.5-Decide at 42.5, Bosun v3.1 0.6B at 38.5, Decision 1.0 Kai at 35.9 and Decision 1.0 Lex at 31.0 JevArena; competitor rows are a public board snapshot dated 2026-09-28.
Reactions & coverage
Posts, threads, and videos about this entry from around the web.
Hacker News: ASTGL replay chose local Laya routing over hosted Jev on one Mac Studio setup.
Jmeg8r15 pointsDoes an open-weight decision model beat a hosted one? Jev vs. LayaRead the thread on Hacker NewsReddit: A walkthrough of routing between models with Jev.
X: Standard One SH (StandardThinking)
Facebook: Intake Router keeps diagnosis out of the answer space at the front desk.
FacebookClinicnotes
More Jev resources like this
The 9B entry in vLLM Semantic Router's Decision 2.0 family: a 7.94B decision model with a 16,384-token context that answers choice, yes/no and score questions in one pass, reporting 68.1 JevArena and 18.4 ms median single-question latency.
Jev Models & Reimplementations#community#open-models#local
Community
The 4B model in vLLM Semantic Router's Decision 2.0 family: a 4.21B fine-tune of Qwen3.5-4B-Base that answers choice, yes/no and score questions in one forward pass, with a self-reported JevArena 63.6 and a 12.9 ms median per request.
Jev Models & Reimplementations#community#open-models#local
Community
The 2B model in vLLM Semantic Router's Decision 2.0 family: a 1.88B fine-tune of the earlier Decision-1.0-Sol-2B checkpoint with a 16,384-token context, reporting JevArena 52.1 and a 7.2 ms median per request.
Jev Models & Reimplementations#community#open-models#local
Community
The 0.8B model in vLLM Semantic Router's Decision 2.0 family: a 0.75B fine-tune of Decision-1.0-Eos-0.8B with a 16,384-token context, reporting JevArena 53.9 and a 6.0 ms median per single-question request.
Jev Models & Reimplementations#community#open-models#local
Community
The vLLM Semantic Router project's 29.37B decision model on Qwen3.8-27B: a probability per supplied option with zero generated tokens, 74.0 on JevArena and 56.5 on Jev Decision Index 0.2.1.
Jev Models & Reimplementations#community#open-models#local
Community
A small distilled sibling of Perplexity's pplx-decider-v1.2: a Qwen3.5-4B decider that reads a state plus typed choice, noul and score questions in one pass, keeping most of the 27B model's Decision Index.
Jev Models & Reimplementations#community#open-models#multimodal
Community
From the Jev guides
Original write-ups on what this entry is for.
From the Jev AI model community
Posts from builders shipping with Jev right now.