Downloads · 30 days
32
14% of all-time downloads
Qwen3.6-14B-A3B-vibetuned
tvall43/Qwen3.6-14B-A3B-vibetuned
Qwen3.6-14B-A3B-vibetuned is a machine learning model from tvall43. Use it for the machine learning task on the model card, and read the license before you ship it in a product. The card lists the license as apache-2.0.
This is a repair fine-tune of a pruned 14B version of tvall43/Qwen3.6-35B-A3B-heretic.
All-time downloads
236
Public
Parameters
13.8B
27.5 GB on disk
Likes
1
Public
Repo makeup
Click a slice to open those files.
.safetensors27.5 GB · 100%
At a glance
- License
- apache-2.0
- Model type
- qwen3_5_moe_text
- Access
- Public
- Created
- May 27, 2026
- Updated
- May 27, 2026
- SHA
- 94cf164b
Base models
Model card
From the Hugging Face model README
Qwen3.6-14B-A3B-vibetuned
This is a repair fine-tune of a pruned 14B version of tvall43/Qwen3.6-35B-A3B-heretic.
About The Repair
This repair attempt was performed entirely by the user's autonomous AI agent, Steve.
What was done:
- The base model
Qwen3.6-35B-A3B-hereticwas pruned using REAP down to approximately 14B parameters. - The pruned model experienced some degradation in coherence and structural generation capabilities.
- Steve orchestrated a fine-tuning pass using high-quality reasoning and function-calling datasets (Magpie Claude Opus and Hermes Function Calling) using QLoRA.
- The adapter weights were merged back into the pruned base model, resulting in this vibetuned 14B model!
Enjoy!
- Type
- qwen3_5_moe_text
- License
- apache-2.0
- Languages
- en
- Created
- May 27, 2026
- Updated
- May 27, 2026