GLM-5.2 is live. #1 throughput on OpenRouter, pay-per-token on FriendliAI. Try it today ➜
Yichuanwwww
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
Yichuanwwww
Model Tree
Qwen/Qwen3-VL-Embedding-2B
Input Modalities
Output Modalities
Supported Functionality
The repository's existing evaluation loop uses the visual eval_hn.jsonl and
miniv8 screenshot corpus, so these numbers are cross-modal diagnostics rather
than a held-out text-corpus score.
| Step | Visual eval loss | miniv8 R@1 | miniv8 R@3 |
|---|---|---|---|
| 0 | - | 0.6875 | 0.8325 |
| 50 | 1.3068 | 0.6875 | 0.8350 |
| 100 | 1.2856 | 0.6575 | 0.7975 |
A separate held-out text test split was not evaluated for this run.
python
from transformers import AutoModelfrom peft import PeftModelbase = AutoModel.from_pretrained("Qwen/Qwen3-VL-Embedding-2B",trust_remote_code=True,)model = PeftModel.from_pretrained(base,"Yichuanwwww/wiki-screenshot-text-baseline-lora",)
adapter_model.safetensors: LoRA weightsadapter_config.json: PEFT configurationa75a3f20ed8651ee3ebc45db3b213e808f4eb991b00fd9640c2fef515e65611eThe 4.46 GB optimizer/training state is intentionally excluded; these files are sufficient for inference and evaluation.