Quantization made by Richard Erkhov.
Welcome to our Code Model repository! Our model is specifically fine-tuned for code generation tasks. Bud Millenial Code Gen open-source models are currently the State of the Art (SOTA) for code generation, beating all the existing models of all sizes. We have achieved a HumanEval value of 80.48 @ Pass 1, beating proprietary models like Gemini Ultra, Claude, GPT-3.5 etc. by a large margin, and on par with GPT-4 (HumanEval ~ 82. Ref. WizardCoder). Our proprietary model (Bud Code Jr) beats GPT-4 as well with a HumanEval value of 88.2 & a context size of 168K, we will be releasing an API for Researchers, Enterprises, and potential Partners by January 2024 end. If interested, please reach out to
jithinvg@bud.studioFor the millenial models, the eval script in the github repo is used for the above result.
Note: The humaneval values of other models are taken from the official repos of
WizardCoder,
DeepseekCoder,
Gemini etc.
1import torch
2from transformers import AutoTokenizer, AutoModelForCausalLM
3
4tokenizer = AutoTokenizer.from_pretrained("budecosystem/code-millenials-13b")
5model = AutoModelForCausalLM.from_pretrained("budecosystem/code-millenials-13b")
6
7template = """A chat between a curious user and an artificial intelligence assistant. The assistant gives helpful, detailed, and polite answers to the user's questions.
8### Instruction: {instruction} ### Response:"""
9
10instruction = <Your code instruction here>
11
12prompt = template.format(instruction=instruction)
13
14inputs = tokenizer(prompt, return_tensors="pt")
15sample = model.generate(**inputs, max_length=128)
16print(tokenizer.decode(sample[0]))
17
The model is trained of 8 A100 80GB for approximately 15hrs.
