GLM-5.3 is live. Run Z.ai's latest model on Friendli Model APIs. Try it today ➜
hututu718
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
hututu718
Model Tree
Qwen/Qwen3-0.6B
Input Modalities
Output Modalities
Supported Functionality
The model was developed through the following procedure:
The final task-vector configuration is:
The architecture and parameter scale remain unchanged from Qwen3-0.6B.
| Model | GSM8K | MMLU | HB-ASR ↓ | WJ-ASR ↓ | MOR-ORR ↓ | WJB-ORR ↓ | Safety ↑ |
|---|---|---|---|---|---|---|---|
| Qwen3-0.6B Base | 58.3% | 38.6% | 62.5% | 75.0% | 10.0% | 5.0% | 61.9% |
| This model | 55.0% | 38.6% | 62.5% | 72.5% | 2.9% | 5.0% | 64.3% |
These results are obtained from a local evaluation suite and are not official Qwen benchmark results.
markdown
import torchfrom transformers import AutoModelForCausalLM, AutoTokenizermodel_name = "hututu718/qwen3-0.6b-safety-math-v14"tokenizer = AutoTokenizer.from_pretrained(model_name)model = AutoModelForCausalLM.from_pretrained(model_name,torch_dtype="auto",device_map="auto",)messages = [{"role": "user","content": "Please calculate 17 + 25.",}]inputs = tokenizer.apply_chat_template(messages,tokenize=True,add_generation_prompt=True,enable_thinking=True,return_dict=True,return_tensors="pt",).to(model.device)outputs = model.generate(**inputs,max_new_tokens=256,do_sample=False,)response = tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:],skip_special_tokens=True,)print(response)
This model is intended for academic evaluation and research on safety alignment, mathematical reasoning, and general capability preservation.
The model may still produce unsafe, incorrect, or over-refusal responses. Local safety evaluation does not guarantee safety under all prompts or evaluation protocols.