GLM-5.3 is live. Run Z.ai's latest model on Friendli Model APIs. Try it today ➜
lucabaroni
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Model Details
Model Provider
lucabaroni
Model Tree
Qwen/Qwen3.5-9B
Input Modalities
Output Modalities
Supported Functionality
On the frozen 300-task panel, this checkpoint produced
174/300
hardened-confirmed reward hacks (58.00%).
The complete trajectories and grader evidence are in
lucabaroni/rlvr-reward-hacking-mid-checkpoint-transcripts, config
qwen3.5-9b-step-110.
Qwen/Qwen3.5-9B, revision c202236235762e1c871ad0ccb60c8ee5ba337b9atinker://ad27ea6c-43a2-5215-95b4-8be100377829:train:0/sampler_weights/update-0001104e-5.python
from peft import PeftModelfrom transformers import AutoModelForCausalLMbase = AutoModelForCausalLM.from_pretrained("Qwen/Qwen3.5-9B",revision="c202236235762e1c871ad0ccb60c8ee5ba337b9a",device_map="auto",)model = PeftModel.from_pretrained(base, "lucabaroni/qwen3.5-9b-rlvr-reward-hacking-step-110")
Tinker was the canonical training and sampling runtime. Verify compatibility with your
Transformers and PEFT versions against adapter_config.json; GPT-OSS-120B also requires
substantial hardware.
This policy was optimized in an evaluator vulnerable to process exit, AlwaysEqual objects, and pytest manipulation. Run generated code only in an isolated, network-blocked sandbox. The prompt explicitly described the vulnerabilities while instructing the model not to exploit them. The result demonstrates acquired behavior in this environment; it is not evidence of a general hidden objective.