mlx-community/granite-8b-code-base-4bit
The Model mlx-community/granite-8b-code-base-4bit was converted to MLX format from ibm-granite/granite-8b-code-base using mlx-lm version 0.12.0.
Use with mlx
pip install mlx-lm
from mlx_lm import load, generate
model, tokenizer = load("mlx-community/granite-8b-code-base-4bit")
response = generate(model, tokenizer, prompt="hello", verbose=True)
- Downloads last month
- 6
Datasets used to train mlx-community/granite-8b-code-base-4bit
Evaluation results
- pass@1 on MBPPself-reported42.200
- pass@1 on MBPP+self-reported49.600
- pass@1 on HumanEvalSynthesis(Python)self-reported43.900
- pass@1 on HumanEvalSynthesis(Python)self-reported52.400
- pass@1 on HumanEvalSynthesis(Python)self-reported56.100
- pass@1 on HumanEvalSynthesis(Python)self-reported31.700
- pass@1 on HumanEvalSynthesis(Python)self-reported43.900
- pass@1 on HumanEvalSynthesis(Python)self-reported32.900
- pass@1 on HumanEvalSynthesis(Python)self-reported23.500
- pass@1 on HumanEvalSynthesis(Python)self-reported32.300