GLM-5.2 is live. #1 throughput on OpenRouter, pay-per-token on FriendliAI. Try it today ➜
Steven668866
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
Steven668866
Model Tree
Qwen/Qwen2.5-VL-7B-Instruct
Input Modalities
Output Modalities
Supported Functionality
| Metric | Base Qwen2.5-VL-7B | This LoRA |
|---|---|---|
| acc@0.5 | 78.8% (*) | 91.2% |
| acc@0.75 | — | 76.8% |
| mIoU | — | 80.5% |
| box-format validity | 0% (**) | 100% |
| latency / query (Ascend 910B2) | — | ~1.62 s |
(*) base parsed fairly from its native bbox_2d JSON output and rescaled to [0,1000]. (**) base does not emit the target [0,1000] (x1,y1),(x2,y2) format; LoRA does, 100% of the time.
Constructed object-presence-history benchmark testing temporal referring expressions ("the object that was there earlier"):
| condition | acc@0.5 |
|---|---|
| with Temporal Memory log | 86.1% |
| without memory (single frame) | 20.3% |
The 66-point gap shows temporal references require an explicit memory module. This benchmark is synthetic — it demonstrates the mechanism, not natural-video performance.
markdown
from transformers import AutoModelForImageTextToTextfrom peft import PeftModelbase = AutoModelForImageTextToText.from_pretrained("Qwen/Qwen2.5-VL-7B-Instruct", dtype="bfloat16")model = PeftModel.from_pretrained(base, "Steven668866/URIS-Qwen2.5-VL-7B-RefCOCO-LoRA")