chancharikm/all_sft_formats_balanced_20260222_ep6_lr3e5_qwen3-vl-8b

VISIONConcurrent Unit Cost:1Model Size:8BQuant:FP8Context Size:32kTool Calling:SupportedPublished:Mar 29, 2026License:apache-2.0Architecture:Transformer Open Weights Featherless Exclusive Cold

The chancharikm/all_sft_formats_balanced_20260222_ep6_lr3e5_qwen3-vl-8b model is a fine-tuned 8 billion parameter Qwen3-VL language model. It was trained on the all_sft_formats_balanced_20260222_1240 dataset, building upon a previous iteration. This model is designed for general language understanding and generation tasks, leveraging its 32768 token context length for processing extensive inputs.

Loading preview...

Overview

This model, chancharikm/all_sft_formats_balanced_20260222_ep6_lr3e5_qwen3-vl-8b, is a fine-tuned version of an existing Qwen3-VL 8 billion parameter model. It was specifically trained on the all_sft_formats_balanced_20260222_1240 dataset, indicating a focus on diverse supervised fine-tuning (SFT) formats. The training process involved a learning rate of 1e-07, a total batch size of 128 across 8 GPUs, and utilized the AdamW optimizer with a cosine learning rate scheduler over 3 epochs.

Key Training Details

  • Base Model: Qwen3-VL 8B
  • Dataset: all_sft_formats_balanced_20260222_1240
  • Learning Rate: 1e-07
  • Optimizer: AdamW with betas=(0.9, 0.999)
  • Epochs: 3.0
  • Total Batch Size: 128 (across 8 GPUs with 2 gradient accumulation steps)

Intended Uses

While specific intended uses and limitations are not detailed in the provided model card, its fine-tuning on a balanced SFT dataset suggests applicability for a range of instruction-following and conversational tasks. Developers can leverage its 8 billion parameters and 32768 token context for applications requiring robust language understanding and generation.