RichardErkhov/Q-bert_-_Optimus-7B-gguf

Quantization made by Richard Erkhov.

Github

Discord

Request more models

Optimus-7B - GGUF

• Model creator: https://huggingface.co/Q-bert/
• Original model: https://huggingface.co/Q-bert/Optimus-7B/

Name	Quant method	Size
Optimus-7B.Q2_K.gguf	Q2_K	2.53GB
Optimus-7B.IQ3_XS.gguf	IQ3_XS	2.81GB
Optimus-7B.IQ3_S.gguf	IQ3_S	2.96GB
Optimus-7B.Q3_K_S.gguf	Q3_K_S	2.95GB
Optimus-7B.IQ3_M.gguf	IQ3_M	3.06GB
Optimus-7B.Q3_K.gguf	Q3_K	3.28GB
Optimus-7B.Q3_K_M.gguf	Q3_K_M	3.28GB
Optimus-7B.Q3_K_L.gguf	Q3_K_L	3.56GB
Optimus-7B.IQ4_XS.gguf	IQ4_XS	3.67GB
Optimus-7B.Q4_0.gguf	Q4_0	3.83GB
Optimus-7B.IQ4_NL.gguf	IQ4_NL	3.87GB
Optimus-7B.Q4_K_S.gguf	Q4_K_S	3.86GB
Optimus-7B.Q4_K.gguf	Q4_K	4.07GB
Optimus-7B.Q4_K_M.gguf	Q4_K_M	4.07GB
Optimus-7B.Q4_1.gguf	Q4_1	4.24GB
Optimus-7B.Q5_0.gguf	Q5_0	4.65GB
Optimus-7B.Q5_K_S.gguf	Q5_K_S	4.65GB
Optimus-7B.Q5_K.gguf	Q5_K	4.78GB
Optimus-7B.Q5_K_M.gguf	Q5_K_M	4.78GB
Optimus-7B.Q5_1.gguf	Q5_1	5.07GB
Optimus-7B.Q6_K.gguf	Q6_K	5.53GB
Optimus-7B.Q8_0.gguf	Q8_0	7.17GB

Original model description:

license: apache-2.0 datasets: - meta-math/MetaMathQA language: - en pipeline_tag: text-generation tags: - Math

Optimus-7B

Fine-tuned On mistralai/Mistral-7B-v0.1 with meta-math/MetaMathQA

You can use ChatML format.

Open LLM Leaderboard Evaluation Results

Detailed results can be found Here

Metric	Value
Avg.	69.09
ARC (25-shot)	65.44
HellaSwag (10-shot)	85.41
MMLU (5-shot)	63.61
TruthfulQA (0-shot)	55.79
Winogrande (5-shot)	78.77
GSM8K (5-shot)	65.50