mlx-community/CalmeRys-78B-Orpo-v0.1-4bit
The Model mlx-community/CalmeRys-78B-Orpo-v0.1-4bit was converted to MLX format from dfurman/CalmeRys-78B-Orpo-v0.1 using mlx-lm version 0.19.1.
Use with mlx
pip install mlx-lm
from mlx_lm import load, generate
model, tokenizer = load("mlx-community/CalmeRys-78B-Orpo-v0.1-4bit")
prompt="hello"
if hasattr(tokenizer, "apply_chat_template") and tokenizer.chat_template is not None:
messages = [{"role": "user", "content": prompt}]
prompt = tokenizer.apply_chat_template(
messages, tokenize=False, add_generation_prompt=True
)
response = generate(model, tokenizer, prompt=prompt, verbose=True)
- Downloads last month
- 31
Model tree for mlx-community/CalmeRys-78B-Orpo-v0.1-4bit
Base model
dnhkng/RYS-XLarge
Finetuned
MaziyarPanahi/calme-2.1-rys-78b
Finetuned
MaziyarPanahi/calme-2.4-rys-78b
Finetuned
dfurman/CalmeRys-78B-Orpo-v0.1
Dataset used to train mlx-community/CalmeRys-78B-Orpo-v0.1-4bit
Evaluation results
- strict accuracy on IFEval (0-Shot)Open LLM Leaderboard81.630
- normalized accuracy on BBH (3-Shot)Open LLM Leaderboard61.920
- exact match on MATH Lvl 5 (4-Shot)Open LLM Leaderboard37.920
- acc_norm on GPQA (0-shot)Open LLM Leaderboard20.020
- acc_norm on MuSR (0-shot)Open LLM Leaderboard36.370
- accuracy on MMLU-PRO (5-shot)test set Open LLM Leaderboard66.800