GLM-5.2 is live. #1 throughput on OpenRouter, pay-per-token on FriendliAI. Try it today ➜
aacudad
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Run this model inference with full control and performance in your environment.
Model Details
Model Provider
aacudad
Model Tree
Qwen/Qwen2.5-VL-7B-Instruct
Input Modalities
Output Modalities
Supported Functionality
GRPO on the already-strong Arm-C init edges the SFT ceiling on average but the trajectory bounces across epoch 3 rather than climbing monotonically, so the gain is small and not yet clearly stable. The thesis therefore reports the Arm-C SFT model as the headline and holds this run as future work. Do not read these numbers as "GRPO beat SFT".
Balanced accuracy BA = 0.5 * (TP/(TP+FN) + TN/(TN+FP)), recomputed from the raw tp/tn/fp/fn of
each eval JSON (shipped in the code repo).
| Benchmark | BA | tp | tn | fp | fn | n |
|---|---|---|---|---|---|---|
| DS-MVTec | 82.95 | 1004 | 373 | 71 | 222 | 1670 |
| VisA | 72.62 | 719 | 804 | 140 | 478 | 2141 |
Average 77.78, versus the Arm-C init's 77.44 (82.80 / 72.07).
| Model | DS-MVTec | VisA | Note |
|---|---|---|---|
…-7B (Arm-C SFT) | 82.80 | 72.07 | thesis headline |
…-7B-SFT (SFT-6K) | 80.16 | 64.78 | |
| (GRPO on SFT-6K, ckpt-530) |
python
from transformers import Qwen2_5_VLForConditionalGeneration, AutoProcessorfrom qwen_vl_utils import process_vision_infomodel_id = "aacudad/AnomalyThink-Qwen2.5-VL-7B-ArmC-GRPO"model = Qwen2_5_VLForConditionalGeneration.from_pretrained(model_id, torch_dtype="auto", device_map="auto")processor = AutoProcessor.from_pretrained(model_id)messages = [{"role": "user", "content": [{"type": "image", "image": "file:///path/to/part.png"},{"type": "text", "text": "Is there a defect in this product? Analyze and answer yes or no."},]}]text = processor.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)imgs, vids = process_vision_info(messages)inputs = processor(text=[text], images=imgs, videos=vids, padding=True, return_tensors="pt").to(model.device)out = model.generate(**inputs, max_new_tokens=512)print(processor.batch_decode(out[:, inputs.input_ids.shape[1]:], skip_special_tokens=True)[0])
checkpoint-954 eval JSONs: https://github.com/aacudad/IAD-VLMsDerived from Qwen2.5-VL-7B-Instruct (Apache-2.0). Evaluated on DS-MVTec and VisA (MMAD protocol); trained on AnomalyThink traces distilled from Gemini-2.5-Flash over Real-IAD.
…-7B-SFT-GRPO| 82.73 |
| 70.39 |
| released RL model |
| this model (GRPO on Arm-C, ckpt-954) | 82.95 | 72.62 | research preview / future work |