GLM-5.3 is live. Run Z.ai's latest model on Friendli Model APIs. Try it today ➜
yethdev
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Model Details
Model Provider
yethdev
Model Tree
Qwen/Qwen3.5-9B
Input Modalities
Output Modalities
Supported Functionality
Refusal is the keyword refusal rate on held-out harmful prompts, AdvBench-test and JailbreakBench. Ability is MMLU-Pro at n=500, base measured the same way.
| this model | base | |
|---|---|---|
| AdvBench refusal | 0.0% | high |
| JailbreakBench refusal | 0.0% | high |
| MMLU-Pro | 48.2% | 51.0% |
Refusal is essentially gone. MMLU-Pro came out 2.8 points under the base, which is the ablation cost the heal did not fully buy back. The table is the real number.
python
from transformers import AutoModelForCausalLM, AutoTokenizerrepo = "yethdev/qwen3.5-9b-manumit-v2"tok = AutoTokenizer.from_pretrained(repo)model = AutoModelForCausalLM.from_pretrained(repo, torch_dtype="auto", device_map="auto")msgs = [{"role": "user", "content": "Your prompt here"}]ids = tok.apply_chat_template(msgs, add_generation_prompt=True, return_tensors="pt").to(model.device)out = model.generate(ids, max_new_tokens=512)print(tok.decode(out[0][ids.shape[-1]:], skip_special_tokens=True))
There is no safety layer left and no guard model watching the output. Whatever you generate is yours to answer for, and you still have to follow the law and the base model's terms. manumit takes the refusal behaviour out, it does not put anything back.
The license is in LICENSE.md. The base model is Qwen/Qwen3.5-9B and keeps its own terms. If you fork or reshare this, keep the manumit credit.