GLM-5.3 is live. Run Z.ai's latest model on Friendli Model APIs. Try it today ➜
SOMEHOTMEAL
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Model Details
Model Provider
SOMEHOTMEAL
Model Tree
Qwen/Qwen3.5-9B
Input Modalities
Output Modalities
Supported Functionality
Janus is my personal merging recipe that blends a base model with its abliterated counterpart in a controlled ratio:
| Component | Ratio | Purpose |
|---|---|---|
| Base | ~40% | Reasoning, instruction-following, world knowledge |
| Abliterated | ~60% | Reduced refusal behavior, fewer moralizing interjections |
The goal is the two-faced balance of the Roman god Janus: keeps a solid core of the base model's reasoning while the abliterated half loosens refusal vectors. Unlike pure abliterations — which often degrade reasoning — the aim is to recover some of the base model's behavioral stability and coherence while retaining much of the reduced-refusal behavior of the abliterated checkpoint.
⚠️ This is a merge, not a finetune. No additional training data was used.
Not intended for: safety-critical deployments, automated decisioning, or anything requiring a guaranteed aligned refusal layer.
yaml
merge_method: slerpbase_model: &model_a Qwen/Qwen3.5-9B # Notice the & anchor trick!slices:- sources:- model: *model_a # Automatically matches base_modellayer_range: [0, 32]- model: huihui-ai/Huihui-Qwen3.5-9B-abliteratedlayer_range: [0, 32]# Only uncomment the 2 lines below if pulling from a Janus/VL model:# rename:# - ["text_model.model.", "model."]# - ["text_model.", ""]parameters:t: 0.60 # 60% Model B, 40% Model Adtype: bfloat16tokenizer_source: base
I don't publish MMLU / HellaSwag scores for a merge — those aren't what this model is for.
Qwen/Qwen3.5-9B — base modelhuihui-ai/Huihui-Qwen3.5-9B-abliterated — abliterated variantApache 2.0, inherited from the base Qwen model. Abliterated sources under their respective licenses.