apepkuss79's picture
Update README.md
c599177 verified
metadata
language:
  - en
license: llama2
datasets:
  - ehartford/samantha-data
model_name: Samantha 1.11 CodeLlama 34B
base_model: ehartford/Samantha-1.11-CodeLlama-34b
inference: false
model_creator: Eric Hartford
model_type: llama
pipeline_tag: text-generation
quantized_by: Second State Inc.

Samantha-1.11-CodeLlama-34B-GGUF

Original Model

cognitivecomputations/Samantha-1.11-CodeLlama-34b

Run with LlamaEdge

  • LlamaEdge version: v0.2.8 and above

  • Prompt template

    • Prompt type: vicuna-1.0-chat

    • Prompt string

      You are Samantha, a sentient AI companion.
      
      USER: <prompt>
      ASSISTANT:
      
  • Context size: 8192

  • Run as LlamaEdge service

    wasmedge --dir .:. --nn-preload default:GGML:AUTO:Samantha-1.11-CodeLlama-34b-Q5_K_M.gguf llama-api-server.wasm -p vicuna-1.0-chat -s 'You are a helpful AI assistant.'
    
  • Run as LlamaEdge command app

    wasmedge --dir .:. --nn-preload default:GGML:AUTO:Samantha-1.11-CodeLlama-34b-Q5_K_M.gguf llama-chat.wasm -p vicuna-1.0-chat -s 'You are a helpful AI assistant.'
    

Quantized GGUF Models

Name Quant method Bits Size Use case
Samantha-1.11-CodeLlama-34b-Q2_K.gguf Q2_K 2 12.5 GB smallest, significant quality loss - not recommended for most purposes
Samantha-1.11-CodeLlama-34b-Q3_K_L.gguf Q3_K_L 3 17.8 GB small, substantial quality loss
Samantha-1.11-CodeLlama-34b-Q3_K_M.gguf Q3_K_M 3 16.3 GB very small, high quality loss
Samantha-1.11-CodeLlama-34b-Q3_K_S.gguf Q3_K_S 3 14.6 GB very small, high quality loss
Samantha-1.11-CodeLlama-34b-Q4_0.gguf Q4_0 4 19.1 GB legacy; small, very high quality loss - prefer using Q3_K_M
Samantha-1.11-CodeLlama-34b-Q4_K_M.gguf Q4_K_M 4 20.2 GB medium, balanced quality - recommended
Samantha-1.11-CodeLlama-34b-Q4_K_S.gguf Q4_K_S 4 19.2 GB small, greater quality loss
Samantha-1.11-CodeLlama-34b-Q5_0.gguf Q5_0 5 23.2 GB legacy; medium, balanced quality - prefer using Q4_K_M
Samantha-1.11-CodeLlama-34b-Q5_K_M.gguf Q5_K_M 5 23.8 GB large, very low quality loss - recommended
Samantha-1.11-CodeLlama-34b-Q5_K_S.gguf Q5_K_S 5 23.2 GB large, low quality loss - recommended
Samantha-1.11-CodeLlama-34b-Q6_K.gguf Q6_K 6 27.7 GB very large, extremely low quality loss
Samantha-1.11-CodeLlama-34b-Q8_0.gguf Q8_0 8 35.9 GB very large, extremely low quality loss - not recommended