LLM EXPLORER 65,121 MODELS INDEXED

SyncPro SFT by EurekaTian

By EurekaTian · 39 downloads

SyncPro SFT is an open-source language model by EurekaTian. Features: 4b LLM, VRAM: 10.7GB, License: apache-2.0, LLM Explorer Score: 0.28.

Base model:finetune:qwen/qwen3...   Base model:qwen/qwen3.5-4b   Conversational Dataset:eurekatian/syncpro-189...   Dual-stream   En   Image-text-to-text   Multimodal   Proactive-video-understanding   Qwen3 5   Region:us   Safetensors   Sharded   Streaming-video   Syncpro   Tensorflow   Video
Model Card on HF πŸ€—: https://huggingface.co/EurekaTian/SyncPro-SFT 

SyncPro SFT Parameters and Internals

LLM NameSyncPro SFT
Repository πŸ€—https://huggingface.co/EurekaTian/SyncPro-SFT 
Base Model(s)  Qwen/Qwen3.5-4B   Qwen/Qwen3.5-4B
Model Size4b
Required VRAM10.7 GB
Updated2026-10-21
MaintainerEurekaTian
Model Typeqwen3_5
Model Files  5.0 GB: 1-of-3   5.0 GB: 2-of-3   0.7 GB: 3-of-3
Supported Languagesen
Model ArchitectureQwen3_5ForConditionalGeneration
Licenseapache-2.0
Model Max Length262144
Transformers Version5.3.0
Tokenizer ClassTokenizersBackend
Padding Token<|endoftext|>
Errorsreplace

Best Alternatives to SyncPro SFT

Best Alternatives
Context / RAM
Downloads
Likes
AgentHorse 4B256K / 8.4 GB9737
Qwen3.5 4B0K / 9.3 GB6575502801
LightOnOCR 3 4B0K / 9.1 GB239571
NuExtract30K / 0.2 GB441570324
Agents A1 4B0K / 9.1 GB17757872
UserRepGen 4B0K / 9.1 GB100813
FrogNano 4B 26090K / 9.3 GB733113
Qwen3.5 4B Base0K / 9.3 GB22781479
Intern Decision 4B0K / 0.1 GB33848
Mizan V3 3B0K / 0.1 GB13518
Note: green Score (e.g. "73.2") means that the model is better than EurekaTian/SyncPro-SFT.