mlx-community/granite-3b-code-base-8bit
The Model mlx-community/granite-3b-code-base-8bit was converted to MLX format from ibm-granite/granite-3b-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-3b-code-base-8bit")
response = generate(model, tokenizer, prompt="hello", verbose=True)
- Downloads last month
- 6
Datasets used to train mlx-community/granite-3b-code-base-8bit
Evaluation results
- pass@1 on MBPPself-reported36.000
- pass@1 on MBPP+self-reported45.100
- pass@1 on HumanEvalSynthesis(Python)self-reported36.600
- pass@1 on HumanEvalSynthesis(Python)self-reported37.200
- pass@1 on HumanEvalSynthesis(Python)self-reported40.900
- pass@1 on HumanEvalSynthesis(Python)self-reported26.200
- pass@1 on HumanEvalSynthesis(Python)self-reported35.400
- pass@1 on HumanEvalSynthesis(Python)self-reported22.000
- pass@1 on HumanEvalSynthesis(Python)self-reported25.000
- pass@1 on HumanEvalSynthesis(Python)self-reported18.900
- pass@1 on HumanEvalSynthesis(Python)self-reported29.900
- pass@1 on HumanEvalSynthesis(Python)self-reported17.100
- pass@1 on HumanEvalSynthesis(Python)self-reported26.800
- pass@1 on HumanEvalSynthesis(Python)self-reported14.000
- pass@1 on HumanEvalSynthesis(Python)self-reported18.300
- pass@1 on HumanEvalSynthesis(Python)self-reported23.200
- pass@1 on HumanEvalSynthesis(Python)self-reported29.900
- pass@1 on HumanEvalSynthesis(Python)self-reported24.400
- pass@1 on HumanEvalSynthesis(Python)self-reported16.500
- pass@1 on HumanEvalSynthesis(Python)self-reported3.700