Views
No views yet
| Benchmark | Metric | Score |
|---|---|---|
| HumanEval | Pass@1 | 21.34% |
| MBPP | Pass@1 | 38.7% |
| GSM8K | Accuracy | 65.2% |
| MATH | Accuracy | 45.8% |
| MMLU | Accuracy | 58.3% |
1from transformers import AutoTokenizer, AutoModelForCausalLM
2import torch
3
4model_name = "Arioron/Vex-Amber-Mini-1.2"
5tokenizer = AutoTokenizer.from_pretrained(model_name)
6model = AutoModelForCausalLM.from_pretrained(
7 model_name,
8 torch_dtype=torch.float16,
9 device_map="auto"
10)
11
12# Code generation example
13prompt = "Write a Python function to reverse a linked list:"
14inputs = tokenizer(prompt, return_tensors="pt").to(model.device)
15
16outputs = model.generate(
17 **inputs,
18 max_new_tokens=256,
19 temperature=0.7,
20 do_sample=True,
21 top_p=0.9,
22 pad_token_id=tokenizer.eos_token_id
23)
24
25print(tokenizer.decode(outputs[0], skip_special_tokens=True))1# Example: The model can generate efficient algorithms
2def quick_sort(arr):
3 if len(arr) <= 1:
4 return arr
5 pivot = arr[len(arr) // 2]
6 left = [x for x in arr if x < pivot]
7 middle = [x for x in arr if x == pivot]
8 right = [x for x in arr if x > pivot]
9 return quick_sort(left) + middle + quick_sort(right)1# Example: Solve quadratic equations and explain steps
2"""
3Solve: x² - 5x + 6 = 0
4Step 1: Factor the equation: (x - 2)(x - 3) = 0
5Step 2: Set each factor to zero: x - 2 = 0 or x - 3 = 0
6Step 3: Solve for x: x = 2 or x = 3
7"""1@misc{vexambermini1.2,
2 title = {Vex Amber Mini 1.2: A Compact Language Model for Code and Mathematics},
3 author = {Arioron},
4 year = {2025},
5 publisher = {Hugging Face},
6 howpublished = {\url{https://huggingface.co/Arioron/Vex-Amber-Mini-1.2}}
7}