Uploaded finetuned model

  • Finetuned by: ramachandrajoshi
  • License: apache-2.0 (Ref: Orignial model License)
  • Finetuned from model : unsloth/Qwen3.5-2B

This qwen3_5 model was trained 2x faster with Unsloth and Huggingface's TRL library.

Here is some result with new comparsion:

image

Thanks to NSP for supporting my effort and encouraging the model development.

Downloads last month
93
Safetensors
Model size
2B params
Tensor type
F32
·
BF16
·
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for ramachandrajoshi/Qwen3.5-2B-KN

Finetuned
Qwen/Qwen3.5-2B
Finetuned
(68)
this model

Dataset used to train ramachandrajoshi/Qwen3.5-2B-KN