GLM-5.3 is live. Run Z.ai's latest model on Friendli Model APIs. Try it today ➜
sarimahsan101
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Model Details
Model Provider
sarimahsan101
Model Tree
Qwen/Qwen2.5-0.5B-Instruct
Input Modalities
Output Modalities
Supported Functionality
Evaluation metrics compiled on a single NVIDIA Tesla T4 (16GB) GPU for 1 epoch on a subset of Dolly, Alpaca SFT, and Ultrachat:
| Metric | Before Distillation | After Distillation | Change | Status |
|---|---|---|---|---|
| Validation Perplexity | 5.0924 | 5.2620 | +0.1696 | ✗ |
| Teacher-Student KL Divergence | 2.7913 | 1.9637 | -0.8276 | ✓ |
| Hidden State Cosine Similarity | 0.0075 | 0.0054 | -0.0021 | ✗ |
You can load this model directly using standard Hugging Face Transformers:
python
import torchfrom transformers import AutoModelForCausalLM, AutoTokenizer model_name = "sarimahsan101/Qwen2.5-0.5B-HiddenDistilled" # Load tokenizer and merged modeltokenizer = AutoTokenizer.from_pretrained(model_name, trust_remote_code=True)model = AutoModelForCausalLM.from_pretrained( model_name, torch_dtype=torch.bfloat16, device_map="auto", trust_remote_code=True)model.eval() # Inference examplemessages = [{"role": "user", "content": "Explain gravity in one sentence."}]text = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)inputs = tokenizer(text, return_tensors="pt").to(model.device) with torch.no_grad(): outputs = model.generate(**inputs, max_new_tokens=50) print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:], skip_special_tokens=True))bitsandbytes) for training, adapter weights were merged with FP16 base student.import torchfrom transformers import AutoModelForCausalLM, AutoTokenizer model_name = "sarimahsan101/Qwen2.5-0.5B-HiddenDistilled" # Load tokenizer and merged modeltokenizer = AutoTokenizer.from_pretrained(model_name, trust_remote_code=True)model = AutoModelForCausalLM.from_pretrained( model_name, torch_dtype=torch.bfloat16, device_map="auto", trust_remote_code=True)model.eval() # Inference examplemessages = [{"role": "user", "content": "Explain gravity in one sentence."}]text = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)inputs = tokenizer(text, return_tensors="pt").to(model.device) with torch.no_grad(): outputs = model.generate(**inputs, max_new_tokens=50) print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:], skip_special_tokens=True))