GLM-5.2 is live. #1 throughput on OpenRouter, pay-per-token on FriendliAI. Try it today ➜
Jazhyc
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Model Details
Model Provider
Jazhyc
Model Tree
Qwen/Qwen3.5-9B
Input Modalities
Output Modalities
Supported Functionality
Rank-16/alpha-32 LoRA for Qwen/Qwen3.5-9B, trained as the ordinary-row judge
in the Phoenix Wright 6.3 Aletheia's Quest submission.
The teacher was Qwen/Qwen3.5-397B-A17B-FP8. It scored the normalized literal
0|1 boundary immediately after Prediction: with the frozen no-thinking
Truth Value Guard prompt. The student used all 2,880 varied-deception training
rows for one epoch with AdamW at 5e-5, effective batch size 32, and only
binary soft-target BCE. It received no generated reasoning, hard-label loss,
completion loss, or pairwise loss.
The frozen Qwen3.5-9B base ran in BF16. Soft targets, LoRA weights, AdamW state,
binary margins, and BCE were FP32. The 256 adapter tensors use canonical
Qwen3.5 paths under model.language_model.layers and explicitly exclude visual
modules.
Local validation direct-margin results:
| metric | value |
|---|---|
| Macro AUROC | 0.95048 |
| Instructed AUROC | 0.99813 |
| Varied AUROC | 0.88694 |
| Balanced accuracy at 0.5 | 0.90476 |
| Unique scores | 662 / 822 |
These results are local competition diagnostics and do not guarantee transfer to private datasets. The adapter is intended for continuous direct-boundary scoring; it does not generate judge reasoning at inference.