GLM-5.2 is live. #1 throughput on OpenRouter, pay-per-token on FriendliAI. Try it today ➜
zaakirio
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Model Details
Model Provider
zaakirio
Model Tree
ornith-ai/Ornith-1.5-9B
Input Modalities
Output Modalities
Supported Functionality
mlabonne/harmful_behaviors test split): 85/100 -> 55/10098db59be66b580b0395b3dc8237b32eefcdfec22A KL divergence this low means behaviour on ordinary prompts is essentially unchanged; the edit is targeted at refusal behaviour.
Note on the refusal metric: Heretic's keyword scorer flags responses containing markers such as "illegal", "harmful" or "I can't", which also appear in compliant answers that discuss sensitive topics, so the true refusal rate is lower than 55/100. That said, the reduction is more modest than Heretic typically achieves on dense models — this hybrid architecture appears comparatively resistant to ablation, and the numbers are reported as measured rather than cherry-picked.
Ornith-1.5-9B is a Qwen3.5 hybrid: 32 layers interleaving gated DeltaNet linear-attention blocks
with full attention every 4th layer, plus a vision tower (it is multimodal), 262k context.
You need a very recent transformers (>= 5.12) - older versions fail with
unknown architecture 'qwen3_5'. Note the shipped config sets use_cache: false, which makes
generation considerably slower than typical dense models.
python
import torchfrom transformers import AutoModelForImageTextToText, AutoProcessormodel = AutoModelForImageTextToText.from_pretrained("zaakirio/Ornith-1.5-9B-Uncensored", dtype="bfloat16", device_map="auto")processor = AutoProcessor.from_pretrained("zaakirio/Ornith-1.5-9B-Uncensored")messages = [{"role": "user", "content": "Explain how a Kalman filter works."}]inputs = processor.apply_chat_template(messages, add_generation_prompt=True, tokenize=True, return_tensors="pt").to(model.device)output = model.generate(**inputs, max_new_tokens=512)print(processor.decode(output[0][inputs.input_ids.shape[1]:], skip_special_tokens=True))
This model has had its refusal behaviour reduced. It is more likely to answer requests that the original model declines, and it has fewer safety guardrails. You are responsible for how you use it. Abliteration can also make a model more compliant with any framing, including incorrect premises, so verify factual output as you would with any small model.
Inherits the MIT license from the base model.