Jev directory
CommunityJev Models & Reimplementations
Turnike-JEV-9B Merged
The berkay123001/turnike-jev-9b LoRA adapter merged into Qwen/Qwen3.5-9B in bf16, so the bilingual Turkish and English decision model can be served without PEFT through the karar.py Karar interface.
- Reading time
- 2 min
- Published by
- Community
- Author
- berkay123001
- Last verified
- Oct 2026
- Languages
- Python
- Capabilities
- Open sourceSelf-hostedRuns locally
Highlights
- bf16 merge of the turnike-jev-9b LoRA adapter into Qwen/Qwen3.5-9B, the same merge karar.py performs at load time.
- Runs without PEFT: Karar(None, taban="berkay123001/turnike-jev-9b-merged") answers a text and closed questions with option probabilities.
- Apache-2.0 repository; the base Qwen/Qwen3.5-9B is Apache-2.0, Copyright 2026 Alibaba Cloud.
- The card defers evaluation, training data, limitations and declarations to the turnike-jev-9b adapter card.
Quickstart
python
from karar import Karar # karar.py from berkay123001/turnike-jev-9b
k = Karar(None, taban="berkay123001/turnike-jev-9b-merged")
k.sor("Kulaklık bozuk geldi, paramı geri istiyorum.", {"Yazar iade istiyor mu?": {"type": "noul"}})Watch out
Card reviewed 2026-10-17; this merged-weights card carries no benchmarks or evaluation of its own and points to the turnike-jev-9b adapter card, whose JevBench numbers are self-reported by the author, so treat accuracy claims for this merge as inherited rather than measured here.
Questions about this entry
Short answers drawn from the linked project or page.
The LoRA adapter berkay123001/turnike-jev-9b merged into Qwen/Qwen3.5-9B in bf16. It is the same merge that karar.py performs at load time, published so the bilingual Turkish and English decision model can be loaded without PEFT.
They produce the same decisions: this repository stores the merged bf16 weights, while the adapter repository ships a roughly 173 MB LoRA (r=16) that karar.py merges while loading. Use this one when you want to skip PEFT at load time.
It reads a text plus closed questions and returns option probabilities in a single forward pass, without generating text and without network calls, mirroring the typed-question contract the Turnike family follows.
Download karar.py from berkay123001/turnike-jev-9b, then call k = Karar(None, taban="berkay123001/turnike-jev-9b-merged") and k.sor(text, questions). Passing taban selects this merged repository instead of loading and merging the adapter.
The repository is Apache-2.0, and the base model Qwen/Qwen3.5-9B is also Apache-2.0, Copyright 2026 Alibaba Cloud, as stated on the card.
It points to berkay123001/turnike-jev-9b for the model card, evaluation, training data, limitations and declarations, and notes that the training data is not released, that the model was not distilled from a commercial decision API, and that AI assistance was used.
Qwen/Qwen3.5-9B: the card lists both Qwen/Qwen3.5-9B and berkay123001/turnike-jev-9b as base models with a merge relation, and the merge is the bf16 one karar.py reproduces at load time.
Reactions & coverage
Posts, threads, and videos about this entry from around the web.
X: Standard One SH (StandardThinking)
LinkedIn: What multimodal decision models actually change, and what they do not.
LinkedInJevDirectory.org
Hacker News: Decision Index 0.2.1 scores open reproductions of Jev on a frozen suite.
jackbravo4 pointsDecision Index 0.2.1 – 70 open reproductions of TypeSafe Jev's Decision ModelRead the thread on Hacker NewsHacker News: Pointer to a DiffusionGemma plus vLLM tutorial for a Jev-shaped classifier.
1 pointHow to build a Jev-style classifier with DiffusionGemma and vLLMRead the thread on Hacker NewsMore Jev resources like this
The berkay123001/turnike-jev-4b LoRA adapter merged into Qwen/Qwen3.5-4B in bf16 so it can be served without PEFT: a bilingual Turkish and English decision model whose parent card reports 0.774 JevBench macro accuracy.
Jev Models & Reimplementations#community#open-models#local
Community
A 9B bilingual Turkish and English Jev-class decision model from Berkay Hasret: a 173 MB LoRA adapter on Qwen/Qwen3.5-9B that answers up to 16 typed questions about one text in a single forward pass and reports 0.783 macro accuracy on full JevBench.
Jev Models & Reimplementations#community#open-models#local
Community
A bilingual Turkish and English 4B Jev-class decision model from Berkay Hasret: a roughly 130 MB LoRA adapter on Qwen/Qwen3.5-4B that answers up to 16 typed questions per text in one pass and reports 0.774 JevBench macro accuracy.
Jev Models & Reimplementations#community#open-models#local
Community
A local yes/no decision model that speaks Jev's wire format on a laptop CPU, refusing Choice and Score until those question types ship.
Jev Models & Reimplementations#community#python#open-models
Community
feder-cr/jevC++
feder-cr/jevC++A Python server in front of llama-server that reads one-token label probabilities and exposes them as POST /v1/systemone.
Jev Models & Reimplementations#community#python#open-models
Community
khimaros/verdictPython
khimaros/verdictPythonA LoRA plus 24-slot decision head over Qwen3.8-27B from Ming Xu's train-jev project: it serves TypeSafe's System One API with 0.851 Kev transfer-v4 accuracy and ECE 0.053, and needs an 80 GB GPU.
Jev Models & Reimplementations#community#open-models#local
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.