Berkesule/Qwen3-VL-8B-Instruct-gemini3pro-tumveri-sft
VISIONConcurrency Cost:1Model Size:8BQuant:FP8Ctx Length:32kPublished:Jan 26, 2026License:apache-2.0Architecture:Transformer Open Weights Cold
Berkesule/Qwen3-VL-8B-Instruct-gemini3pro-tumveri-sft is an 8 billion parameter Qwen3-VL instruction-tuned model developed by Berkesule. This model was fine-tuned using Unsloth and Huggingface's TRL library, resulting in a 2x faster training process. It is designed for instruction-following tasks, leveraging its Qwen3-VL architecture for multimodal capabilities.
Loading preview...
Overview
Berkesule/Qwen3-VL-8B-Instruct-gemini3pro-tumveri-sft is an 8 billion parameter instruction-tuned model based on the Qwen3-VL architecture. Developed by Berkesule, this model was fine-tuned from unsloth/Qwen3-VL-8B-Instruct using the Unsloth library and Huggingface's TRL, which enabled a 2x acceleration in the training process.
Key Capabilities
- Instruction Following: Optimized for understanding and executing instructions.
- Multimodal Architecture: Leverages the Qwen3-VL base for potential vision-language tasks.
- Efficient Training: Benefits from Unsloth's optimizations for faster fine-tuning.
Good For
- Developers seeking an instruction-tuned Qwen3-VL model.
- Applications requiring efficient fine-tuning for specific instruction-following use cases.
- Exploring multimodal capabilities within an 8B parameter model.