matrixportal/layerskip-llama3.2-1B-GGUF

layerskip-llama3.2-1B GGUF Quantized Models

Technical Details

• Quantization Tool: llama.cpp
• Version: version: 5092 (d3bd7193)

Model Information

• Base Model: facebook/layerskip-llama3.2-1B
• Quantized by: matrixportal

Available Files

• layerskip-llama3.2-1b.q2_k.gguf (553.96MB)
• layerskip-llama3.2-1b.q3_k_s.gguf (611.96MB)
• layerskip-llama3.2-1b.q3_k_m.gguf (658.84MB)
• layerskip-llama3.2-1b.q3_k_l.gguf (698.59MB)
• layerskip-llama3.2-1b.q4_0.gguf (735.21MB)
• layerskip-llama3.2-1b.q4_k_s.gguf (739.71MB)
• layerskip-llama3.2-1b.q4_k_m.gguf (770.27MB)
• layerskip-llama3.2-1b.q5_0.gguf (851.21MB)
• layerskip-llama3.2-1b.q5_k_s.gguf (851.21MB)
• layerskip-llama3.2-1b.q5_k_m.gguf (869.27MB)
• layerskip-llama3.2-1b.q6_k.gguf (974.46MB)
• layerskip-llama3.2-1b.q8_0.gguf (1259.88MB)
• layerskip-llama3.2-1b.f16.gguf (2364.72MB)

💡 Q4_K_M provides the best balance for most use cases