GLM-5.2 is live. #1 throughput on OpenRouter, pay-per-token on FriendliAI. Try it today ➜
mudasir13cs
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
mudasir13cs
Model Tree
Qwen/Qwen2.5-VL-3B-Instruct
Input Modalities
Output Modalities
Supported Functionality
Original illustration for the model card (not a CubiCasa training image):
| Resource | Link |
|---|---|
| CubiCasa5K paper | A Dataset and an Approach to Floorplan Reconstruction |
| Zenodo download | doi:10.5281/zenodo.2613548 |
| License | CC BY-NC 4.0 |
| HF mirrors | Claudio9701/cubicasa5k · Forceless/Zenodo10K |
| Method | FloorplanVLM (arXiv:2602.06507) |
| Stage 1 adapter | mudasir13cs/qwen25-vl-3b-floorplan-sft |
GRPO continues from the SFT adapter using geometric rewards (JSON validity, outer-wall IoU, room IoU) on CubiCasa-derived JSON targets.
train_floorplan_grpo.py — reward R = 0.1·R_val + 0.5·R_ext + α·0.4·R_int; SFT_MODEL_ID, HUB_MODEL_ID, and OUTPUT_DIR are edited at the top of that script.If you are working inside a clone of your training repo, the same files may exist locally beside this folder (relative paths).
bash
pip install torch torchvision transformers trl peft accelerate pillow
(Training also uses Shapely, datasets, numpy; inference does not strictly need Shapely.)
Use Hub IDs (recommended):
python
from transformers import Qwen2_5_VLForConditionalGeneration, AutoProcessorfrom peft import PeftModelBASE = "Qwen/Qwen2.5-VL-3B-Instruct"ADAPTER = "mudasir13cs/qwen25-vl-3b-floorplan-grpo"processor = AutoProcessor.from_pretrained(BASE)model = Qwen2_5_VLForConditionalGeneration.from_pretrained(BASE, torch_dtype="auto", device_map="auto")model = PeftModel.from_pretrained(model, ADAPTER)model.eval()
If you cloned weights into this folder locally: set ADAPTER = "./floorplan-vlm-grpo" (or an absolute path) instead of the Hub repo id.
For best alignment with training, reuse the system prompt that includes the full JSON schema from stage 1 (see SYSTEM_PROMPT in train_floorplan_vlm.py). GRPO training uses a shorter system string in train_floorplan_grpo.py; either works, but the schema-explicit SFT prompt below usually yields more stable JSON.
User text (same as both scripts): “Vectorize this floor plan into structured JSON with all walls, doors, windows, and rooms.”
Minimal pattern (aligned with the inference test in train_floorplan_vlm.py):
python
import json, re, torchfrom PIL import Image# Same strings as train_floorplan_vlm.py (schema-in-the-prompt; recommended for decoding).SYSTEM_PROMPT = ("You are a floor plan vectorization expert. Extract wall, door, window geometry ""from floor plan images into structured JSON.\n\n""Output ONLY valid JSON with this schema:\n"'{"walls":[{"id":"wall_N","start":[x,y],"end":[x,y],"thickness":T,"curvature":0,''"openings":[{"type":"door"|"window","center":D,"width":W}]}],''"rooms":[{"label":"room_type","walls":["wall_N",...]}]}\n\n'"Coordinates normalized so longer image edge = 1024.")USER_PROMPT = "Vectorize this floor plan into structured JSON with all walls, doors, windows, and rooms."image = Image.open("plan.png").convert("RGB")messages = [{"role": "system", "content": [{"type": "text", "text": SYSTEM_PROMPT}]},{"role": "user", "content": [{"type": "image"}, {"type": "text", "text": USER_PROMPT}]},]text = processor.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)inputs = processor(text=[text], images=[image], return_tensors="pt", padding=True)inputs = {k: v.to(model.device) if hasattr(v, "to") else v for k, v in inputs.items()}with torch.no_grad():out = model.generate(**inputs, max_new_tokens=4096, do_sample=False)raw = processor.batch_decode(out[:, inputs.input_ids.shape[1] :], skip_special_tokens=True)[0]m = re.search(r"\{[\s\S]*\}", raw)plan = json.loads(m.group()) if m else None
Output shape: top-level walls (with optional openings) and rooms. Example JSON shape is spelled out under Output JSON Schema in the manitocross training README.
mudasir13cs/qwen25-vl-3b-floorplan-sft.train_floorplan_grpo.py from a checkout with the environment described there (CubiCasa5K under ./cubicasa_data, huggingface-cli login if PUSH_TO_HUB = True; config block at top of file).bibtex
@article{floorplanvlm2026,title={FloorplanVLM: A Vision-Language Model for Floorplan Vectorization},journal={arXiv preprint arXiv:2602.06507},year={2026}}
Mudasir — Sr. AI Engineer at ECODA (에코다), building multimodal AI for architecture and building-performance workflows. MS AI Convergence, 숭실대학교 — Soongsil University, Seoul. More credentials, publications, and projects: mudasir13cs.github.io
train_floorplan_vlm.py