xwinlm:7b-v0.2-q2_K

xwinlm:7b-v0.2-q2_K

101.7K Downloads Updated 1 year ago

Conversational model based on Llama 2 that performs competitively on various benchmarks.

7b 13b

Updated 1 year ago

1 year ago

fe7141ff19bb · 2.8GB ·

archllama

·

parameters6.74B

·

quantizationQ2_K

2.8GB

{{ .System }} User: {{ .Prompt }} Assistant:

45B

{ "num_ctx": 4096, "stop": [ "User:", "Assistant:" ] }

46B

Readme

Xwin-LM is a model based on Llama 2 using various techniques to improve its quality.

Reference