Jev directory
CommunityJev Models & Reimplementations
JEV-35B on ModelScope
AutoTrust's JEV-35B page on ModelScope: metadata records a fine-tune of Qwen/Qwen3.5-35B-A3B under the Qwen3_5MoeForConditionalGeneration architecture, apache-2.0, 521 downloads, English only, with safetensors and PyTorch files.
- Reading time
- 2 min
- Published by
- Community
- Author
- AutoTrust
- Last verified
- Oct 2026
- Capabilities
- Open sourceSelf-hostedRuns locally
Highlights
- ModelScope records the base model as Qwen/Qwen3.5-35B-A3B with the relation finetune and architecture Qwen3_5MoeForConditionalGeneration.
- Licence field reads apache-2.0; the page showed 521 downloads with English as the only listed language.
- Ships as safetensors and pytorch libraries from a GitHub upload into the autotrust organisation's namespace.
- The card-ready flag is 0: this capture carries the platform metadata only, not a published model card or readme prose.
- No metrics or dataset entries are attached to this repository in the captured page, unlike its JEV-9B sibling.
Watch out
Only ModelScope's JSON metadata is visible here: the readme, benchmarks, and model card were unavailable, so accuracy, context window, and hardware needs are not established by this source. The apache-2.0 label is a metadata field, and the 35B-A3B mixture-of-experts base implies large GPU requirements, but the page states no serving instructions or memory figures.
Questions about this entry
Short answers drawn from the linked project or page.
A repository under the autotrust organisation, recorded as a fine-tune of Qwen/Qwen3.5-35B-A3B under the Qwen3_5MoeForConditionalGeneration architecture, in the JEV family of open decision models.
Licence apache-2.0, 521 downloads, English as the only listed language, safetensors and pytorch libraries, a pytorch framework entry, upload from GitHub, and a card-ready flag of 0 rather than a published model card.
None is visible. The capture shows an empty Metrics array and an empty Description field, so the page as captured publishes no benchmark numbers or evaluation results for this repository.
The capture does not include a file list or size for this repository; only architecture and format metadata are visible. The download and memory footprint are therefore not established by this page.
The autotrust organisation on ModelScope. Sibling pages in the same ecosystem identify AutoTrust as the organisation, with a linked GitHub account at github.com/AutoTrustAI, which the org description ties to Singapore and Silicon Valley.
Qwen/Qwen3.5-35B-A3B, recorded in the BaseModel field with the relation finetune, while the architecture field names Qwen3_5MoeForConditionalGeneration and the framework entry is pytorch.
The ModelScope metadata records License: apache-2.0. No licence link or licence name text is included in the captured page, and the model card that might spell out terms was not available.
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
AutoTrust's vision-language decision model on ModelScope: an adapter on Qwen3.8-27B, Apache-2.0, 1,735 downloads, with lora, safetensors, and pytorch assets, trained on the SargeDev/jev-distill-corpus-v3 corpus and English metadata.
Jev Models & Reimplementations#community#open-models#local
Community
Bespoke Labs' 9B decision model on Ollama, fine-tuned from Qwen3.5-9B: Apache-2.0, up to 64 typed questions per call, answers in under 100 ms on a MacBook Pro with an M5 Max, and 75.7% mean accuracy across 13 human-labelled datasets.
Jev Models & Reimplementations#community#open-models#local
Community
The typed-decisions checkpoint of the Laya family on ModelScope: a 421M ModernBERT-large fine-tune reporting 0.766 accuracy over 2,000 decisions, above TypeSafe Jev 1.13.0's published 0.727 and the 0.735 teacher ceiling, Apache-2.0.
Jev Models & Reimplementations#community#open-models#local
Community
Docker's verified ai/openjev image: an 18.9 GB artifact converting the OpenJev weights from huggingface.co/openjev/openjev, served with llama serve through the /v1/systemone API and released under CC BY-NC 4.0 for non-commercial use only.
Jev Models & Reimplementations#community#open-models#local
Community
A 3.74B language-diffusion decision model that answers 184 of 231 matching adaptation questions (79.65%) against Jev 1.13.0's public 86.58%, matching it on the easy subset at 33 ms p50 on one RTX 5090.
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++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.