koesn/
llama3-8b-instruct

1,255 1 year ago

Fixed num_ctx to 8192 and eos token. This Llama 3 8B Instruct model is ready to use for full model's 8k contexts window.

5 models

llama3-8b-instruct:latest
c40cd0e317f3 • 4.7GB • 8K context window •
Text input • 1 year ago
llama3-8b-instruct:Q4_0
c40cd0e317f3 • 4.7GB • 8K context window •
Text input • 1 year ago
llama3-8b-instruct:Q4_K_M
aae5b523ef30 • 4.9GB • 8K context window •
Text input • 1 year ago
llama3-8b-instruct:Q5_K_M
7f44284ae112 • 5.7GB • 8K context window •
Text input • 1 year ago
llama3-8b-instruct:Q6_K
a3288dbaf70c • 6.6GB • 8K context window •
Text input • 1 year ago