GLM-5.2 is live. #1 throughput on OpenRouter, pay-per-token on FriendliAI. Try it today ➜
Maorb23
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Model Details
Model Provider
Maorb23
Model Tree
Base
Qwen/Qwen3.6-27B
Adapter
this model
Input Modalities
Text
Image
Video
Output Modalities
Text
Supported Functionality
Dedicated Endpoints
This repository contains LoRA adapter weights only. It does not contain merged or full base-model weights.
Base model: Qwen/Qwen3.6-27B
Training mixture:
json
{"source_counts": {"medmcqa": {"train": 17500,"eval": 350,"weight": 0.35},"pubmedqa": {"train": 15000,"eval": 300,"weight": 0.3},"medqa_usmle": {"train": 10000,"eval": 200,"weight": 0.2},"mednli": {"train": 5000,"eval": 100,"weight": 0.1},"medical_safety": {"train": 2500,"eval": 50,"weight": 0.05}},"weights": {"medmcqa": 0.35,"pubmedqa": 0.3,"medqa_usmle": 0.2,"mednli": 0.1,"medical_safety": 0.05},"train_rows": 50000,"eval_rows": 1000}
Safety scope: medical education and biomedical research reasoning, not direct patient care.