GLM-5.3 is live. Run Z.ai's latest model on Friendli Model APIs. Try it today ➜
sair390
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Model Details
Model Provider
sair390
Model Tree
Input Modalities
Output Modalities
Supported Functionality
| Field | Value |
|---|---|
| Base model | unsloth/qwen2.5-vl-3b-instruct-unsloth-bnb-4bit |
| Architecture | Qwen2.5-VL (3B parameters, vision-language) |
| Fine-tuning technique | QLoRA (4-bit NF4 quantization) |
| Merged | Yes (LoRA weights merged into base) |
| Language | Tamil (ta) |
| Task | OCR / image-to-text |
| Hyperparameter | Value |
|---|---|
| Epochs | 3 |
| Per-device batch size | 1 |
| Gradient accumulation steps | 16 |
| Effective batch size | 16 |
| Learning rate | 2e-4 |
| LR scheduler | Cosine |
| Warmup ratio | 0.03 |
| Weight decay | 0.01 |
| LoRA rank (r) | 64 |
| LoRA alpha | 64 |
| LoRA dropout | 0.05 |
| Max sequence length | 1024 |
| Precision | bf16 |
| Gradient checkpointing | Yes |
| Save / eval steps | 500 |
| Quantization during training | 4-bit NF4, bfloat16 compute |
| Inference temperature | 1.0 (set to 0 for deterministic OCR) |
| Split | Samples | Source |
|---|---|---|
| Train | 159,939 (80%) | Synthetic + real book scans |
| Validation | 19,992 (10%) | Synthetic + real book scans |
| Test | 19,992 (10%) | Synthetic + real book scans |
| Total | 199,925 |
Synthetic data augmentations: blur, noise, rotation, brightness/contrast, JPEG artifacts, erosion, shadow overlays Real data: ~150 Tamil books, ~75K pages Fonts: NotoSerifTamil, NotoSansTamil, MuktaMalar, HindMadurai, Catamaran Text sources: Tirukkural, Wikipedia sentences, digitized book corpus Text layouts: 75% single-line, 25% multi-line Font sizes: 24–72pt (single-line), 24–48pt (multi-line)
python
from transformers import Qwen2_5_VLForConditionalGeneration, AutoProcessorfrom PIL import Imageimport torch model = Qwen2_5_VLForConditionalGeneration.from_pretrained( "sair390/tamil-ocr-qwen25vl", torch_dtype=torch.bfloat16, device_map="auto",)processor = AutoProcessor.from_pretrained("sair390/tamil-ocr-qwen25vl") image = Image.open("tamil_page.jpg")messages = [{ "role": "user", "content": [ {"type": "image", "image": image}, {"type": "text", "text": "Read the Tamil text in this image."}, ],}] text = processor.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)inputs = processor(text=[text], images=[image], return_tensors="pt").to(model.device) with torch.no_grad(): output = model.generate( **inputs, max_new_tokens=512, temperature=0, # deterministic — best for OCR do_sample=False, ) result = processor.decode(output[0], skip_special_tokens=True)print(result)| Mode | VRAM |
|---|---|
| bfloat16 (full precision) | ~12–16 GB |
| 4-bit quantized | ~6–8 GB (use LoRA adapter on quantized base instead) |
Dependencies: transformers, torch, Pillow, qwen-vl-utils
from transformers import Qwen2_5_VLForConditionalGeneration, AutoProcessorfrom PIL import Imageimport torch model = Qwen2_5_VLForConditionalGeneration.from_pretrained( "sair390/tamil-ocr-qwen25vl", torch_dtype=torch.bfloat16, device_map="auto",)processor = AutoProcessor.from_pretrained("sair390/tamil-ocr-qwen25vl") image = Image.open("tamil_page.jpg")messages = [{ "role": "user", "content": [ {"type": "image", "image": image}, {"type": "text", "text": "Read the Tamil text in this image."}, ],}] text = processor.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)inputs = processor(text=[text], images=[image], return_tensors="pt").to(model.device) with torch.no_grad(): output = model.generate( **inputs, max_new_tokens=512, temperature=0, # deterministic — best for OCR do_sample=False, ) result = processor.decode(output[0], skip_special_tokens=True)print(result)