Llama-3-8B-Instruct-262k-GGUF
Original Model
gradientai/Llama-3-8B-Instruct-262k
Run with LlamaEdge
LlamaEdge version: v0.8.3 and above
Prompt template
Prompt type:
llama-3-chat
Prompt string
<|begin_of_text|><|start_header_id|>system<|end_header_id|> {{ system_prompt }}<|eot_id|><|start_header_id|>user<|end_header_id|> {{ user_message_1 }}<|eot_id|><|start_header_id|>assistant<|end_header_id|> {{ model_answer_1 }}<|eot_id|><|start_header_id|>user<|end_header_id|> {{ user_message_2 }}<|eot_id|><|start_header_id|>assistant<|end_header_id|>
Context size:
262144
Run as LlamaEdge service
wasmedge --dir .:. --nn-preload default:GGML:AUTO:Llama-3-8B-Instruct-262k-Q5_K_M.gguf \ llama-api-server.wasm \ --prompt-template llama-3-chat \ --ctx-size 262144 \ --model-name llama-3-8B-instruct-262k
Run as LlamaEdge command app
wasmedge --dir .:. --nn-preload default:GGML:AUTO:Llama-3-8B-Instruct-262k-Q5_K_M.gguf \ llama-chat.wasm \ --prompt-template llama-3-chat \ --ctx-size 262144
Quantized GGUF Models
Name | Quant method | Bits | Size | Use case |
---|---|---|---|---|
Llama-3-8B-Instruct-262k-Q2_K.gguf | Q2_K | 2 | 3.18 GB | smallest, significant quality loss - not recommended for most purposes |
Llama-3-8B-Instruct-262k-Q3_K_L.gguf | Q3_K_L | 3 | 4.32 GB | small, substantial quality loss |
Llama-3-8B-Instruct-262k-Q3_K_M.gguf | Q3_K_M | 3 | 4.02 GB | very small, high quality loss |
Meta-Llama-3-70B-Instruct-Q3_K_S.gguf | Q3_K_S | 3 | 3.66 GB | very small, high quality loss |
Llama-3-8B-Instruct-262k-Q4_0.gguf | Q4_0 | 4 | 4.66 GB | legacy; small, very high quality loss - prefer using Q3_K_M |
Llama-3-8B-Instruct-262k-Q4_K_M.gguf | Q4_K_M | 4 | 4.92 GB | medium, balanced quality - recommended |
Llama-3-8B-Instruct-262k-Q4_K_S.gguf | Q4_K_S | 4 | 4.69 GB | small, greater quality loss |
Llama-3-8B-Instruct-262k-Q5_0.gguf | Q5_0 | 5 | 5.6 GB | legacy; medium, balanced quality - prefer using Q4_K_M |
Llama-3-8B-Instruct-262k-Q5_K_M.gguf | Q5_K_M | 5 | 5.73 GB | large, very low quality loss - recommended |
Llama-3-8B-Instruct-262k-Q5_K_S.gguf | Q5_K_S | 5 | 5.6 GB | large, low quality loss - recommended |
Llama-3-8B-Instruct-262k-Q6_K.gguf | Q6_K | 6 | 6.6 GB | very large, extremely low quality loss |
Llama-3-8B-Instruct-262k-Q8_0.gguf | Q8_0 | 8 | 8.54 GB | very large, extremely low quality loss - not recommended |
Llama-3-8B-Instruct-262k-f16.gguf | f16 | 16 | 16.1 GB |
Quantized with llama.cpp b2734.
- Downloads last month
- 160
Model tree for second-state/Llama-3-8B-Instruct-262k-GGUF
Base model
gradientai/Llama-3-8B-Instruct-262k