LLM EXPLORER 59,962 MODELS INDEXED

Qwen3.8 14B Instruct Turbo by ewinregirgojr

By ewinregirgojr · 2038 downloads
HF Repo ↗ Copy URL

Qwen3.8 14B Instruct Turbo is an open-source language model by ewinregirgojr. Features: 27b LLM, VRAM: 30GB, License: apache-2.0, Instruction-Based, LLM Explorer Score: 0.35.

  14b   Agent Base model:finetune:qwen/qwen3...   Base model:qwen/qwen3.8-27b   Coding   Conversational   Deltanet   Distillation   En   Endpoints compatible   Hybrid-attention   Instruct   Linear-attention   Lora   Pruned   Qwen   Qwen3   Qwen3 5   Reasoning   Region:us   Safetensors   Sharded   Tensorflow   Tool-calling   Zh
Model Card on HF 🤗: https://huggingface.co/ewinregirgojr/Qwen3.8-14B-Instruct-Turbo 

Qwen3.8 14B Instruct Turbo Parameters and Internals

LLM NameQwen3.8 14B Instruct Turbo
Repository 🤗https://huggingface.co/ewinregirgojr/Qwen3.8-14B-Instruct-Turbo 
Base Model(s)  Qwen/Qwen3.8-27B   Qwen/Qwen3.8-27B
Model Size27b
Required VRAM30 GB
Updated2026-08-30
Maintainerewinregirgojr
Instruction-BasedYes
Model Files  0.3 GB   3.0 GB: 1-of-10   3.0 GB: 2-of-10   3.5 GB: 3-of-10   3.0 GB: 4-of-10   3.0 GB: 5-of-10   3.0 GB: 6-of-10   3.0 GB: 7-of-10   3.0 GB: 8-of-10   3.0 GB: 9-of-10   2.5 GB: 10-of-10
Supported Languagesen zh
Model ArchitectureAutoModelForCausalLM
Licenseapache-2.0
Model Max Length262144
Is Biasednone
Tokenizer ClassQwen2Tokenizer
Padding Token<|endoftext|>
PEFT TypeLORA
LoRA ModelYes
PEFT Target Modulesq_proj|down_proj|gate_proj|o_proj|k_proj|up_proj|v_proj
LoRA Alpha64
LoRA Dropout0.05
R Param32
Errorsreplace