LLM EXPLORER 64,206 MODELS INDEXED

Qwen3.5 0.8B Japanese SFT V2.5 by Takenoko12345678

By Takenoko12345678 · 60 downloads
HF Repo ↗ Copy URL

Qwen3.5 0.8B Japanese SFT V2.5 is an open-source language model by Takenoko12345678. Features: 0.8b LLM, VRAM: 1.7GB, License: apache-2.0, Fine-Tuned, LLM Explorer Score: 0.29.

Base model:finetune:qwen/qwen3...   Base model:qwen/qwen3.5-0.8b   Conversational Dataset:nvidia/opensciencereas...   Dataset:openai/gsm8k Dataset:takenoko12345678/japan... Dataset:takenoko12345678/japan... Dataset:takenoko12345678/qwen3...   Dataset:zwhe99/deepmath-103k   En   Endpoints compatible   Finetuned   Full-finetune   Image-text-to-text   Ja   Japanese   Qwen3 5   Reasoning   Region:us   Safetensors
Model Card on HF 🤗: https://huggingface.co/Takenoko12345678/Qwen3.5-0.8B-Japanese-SFT-v2.5 

Qwen3.5 0.8B Japanese SFT V2.5 Parameters and Internals

LLM NameQwen3.5 0.8B Japanese SFT V2.5
Repository 🤗https://huggingface.co/Takenoko12345678/Qwen3.5-0.8B-Japanese-SFT-v2.5 
Base Model(s)  Qwen/Qwen3.5-0.8B   Qwen/Qwen3.5-0.8B
Model Size0.8b
Required VRAM1.7 GB
Updated2026-10-14
MaintainerTakenoko12345678
Model Typeqwen3_5
Model Files  1.7 GB
Supported Languagesja en
Model ArchitectureQwen3_5ForConditionalGeneration
Licenseapache-2.0
Model Max Length262144
Transformers Version5.17.0
Tokenizer ClassQwen2Tokenizer
Padding Token<|endoftext|>
Errorsreplace

Best Alternatives to Qwen3.5 0.8B Japanese SFT V2.5

Best Alternatives
Context / RAM
Downloads
Likes
OvisOCR20K / 1.7 GB265086487
Qwen3.5 0.8B0K / 1.8 GB2839884655
LightOnOCR 3 0.8B0K / 1.7 GB131848
MinerU2.5 Flash 0.8B0K / 1.7 GB231
Qwen3.5 0.8B Base0K / 1.8 GB82805492
Xiaomi OCR 00K / 1.8 GB93315
Tev1 0.8B Onnx Webgpu0K /  GB13361
Tev1 0.8B Experimental0K / 1.8 GB6449
Intern Decision 0.8B0K / 0.2 GB24216
OneJev 0.8B0K / 2.2 GB11012
Note: green Score (e.g. "73.2") means that the model is better than Takenoko12345678/Qwen3.5-0.8B-Japanese-SFT-v2.5.