Skip to content

EthenEthenEthen

LoginPlatformTry Ethen

Open Source Model Profile · vilm

Quyen-Mini-v0.1

Quyen-Mini-v0.1 is a Qwen-family conversational model from vilm trained with SFT and DPO and defaulting to the ChatML template.

Publisher
vilm
Task
text-generation
Model type
qwen2
License
other
Library
transformers
Publication status
Accepted · not indexed

Model overview

Quyen-Mini-v0.1 is published by vilm as a conversational text-generation model in the Quyen series. The captured configuration identifies Qwen2ForCausalLM with model type qwen2. According to the model card, Quyen-Mini is the 1.8B version trained with SFT and DPO, and card data records other as the license.

Recorded capabilities

Qwen-family conversational model

The captured configuration identifies Qwen2ForCausalLM, while the model card describes the Quyen series as based on the Qwen1.5 family with Quyen-Mini as its 1.8B version.

SFT and DPO training mix

According to the model card, training used SFT and DPO across OpenHermes-2.5, Capybara, distilabel-capybara-dpo-7k-binarized, orca_dpo_pairs, and private data.

ChatML default template

According to the model card, all Quyen models use ChatML as the default prompt template.

Use cases in the source record

  • Conversational text-generation work using the publisher's documented ChatML template.
  • Preference-tuned assistant experiments building on the card's documented SFT and DPO workflow.

Limitations and unknowns

  • No parameter count was extracted from this record; the 1.8B figure is a publisher series claim.
  • No evaluation results were extracted from this record.
  • Provider state is historical snapshot data and should be refreshed before being presented as current.

Source and provenance

Source: vilm/Quyen-Mini-v0.1

Captured: Unknown. Processed: 2026-09-07T19:35:00.786466+00:00.

Quyen Model Description Quyen is our first flagship LLM series based on the Qwen1.5 family. We introduced 6 different versions: Quyen-SE (0.5B) Quyen-Mini (1.8B) Quyen (4B) Quyen-Plus (7B) Quyen-Pro (14B) Quyen-Pro-Max (72B) All models were trained with SFT and DPO using the following dataset: OpenHermes-2.5 by Teknium Capyabara by LDJ argilla/distilabel-capybara-dpo-7k-binarized by argilla orca_dpo_pairs by Intel and Private Data by Ontocord & BEE-spoke-data Prompt Template All Quyen models use ChatML as the default template: <|im_start|>system You are a sentient, superintelligent artificial general intelligence, here to teach and…

F001F002F003F004F005F006F008F009F011