Zeta 2.1 - GGUF

Direct GGUF of zed-industries/zeta-2.1, published by bluevoid-pl.

  • Base model: ByteDance-Seed/Seed-Coder-8B-Base
  • Size: 8B params
  • Architecture: llama
  • License: Apache 2.0
  • HF: bluevoid-pl/zeta-2.1-GGUF

Czym is

Zeta 2.1 is a code edit prediction model (next-edit suggestion). Given code context, edit history and an editable region around the cursor, it predicts the rewritten content for that region.

Quantizations prefixed with I use the zed-industries/zeta dataset plus groups_merged.txt for the imatrix.

Running

ollama run hf.co/bluevoid-pl/zeta-2.1-GGUF:Q4_K_M

Quantizations

QuantSize
IQ2_XXS2.51 GB
IQ2_XS2.72 GB
IQ2_S2.88 GB
IQ2_M3.07 GB
Q2_K_S3.11 GB
Q2_K3.30 GB
IQ3_XXS3.40 GB
IQ3_XS3.65 GB
Q3_K_S3.80 GB
IQ3_S3.82 GB
IQ3_M3.92 GB
Q3_K_M4.15 GB
Q3_K_L4.46 GB
IQ4_XS4.59 GB
Q4_04.83 GB
IQ4_NL4.83 GB
Q4_K_S4.84 GB
Q4_K_M5.07 GB
Q8_08.77 GB
BF1616.5 GB