Seed-Coder-8B-Instruct - DASH-Q 2-bit GGUF
llama.cpp GGUF files of ByteDance-Seed/Seed-Coder-8B-Instruct quantized with DASH-Q, at several 2-bit-class sizes. Every file uses only standard llama.cpp tensor types (no tensor above 4 bits) and loads in any recent llama.cpp build.
| File | Type | Size | Bits / weight |
|---|---|---|---|
Seed-Coder-8B-Instruct-DASHQ-IQ2_XXS.gguf |
IQ2_XXS | 2.65 GB | 2.57 |
Seed-Coder-8B-Instruct-DASHQ-IQ2_XS.gguf |
IQ2_XS | 2.98 GB | 2.89 |
Seed-Coder-8B-Instruct-DASHQ-IQ2_M.gguf |
IQ2_M | 3.14 GB | 3.05 |
Seed-Coder-8B-Instruct-DASHQ-Q2_K_XL.gguf |
Q2_K_XL | 3.38 GB | 3.28 |
Perplexity (lower is better)
| Type | Model | Size | WikiText-2 | C4 |
|---|---|---|---|---|
| IQ2_XXS | llama.cpp IQ2_XXS (imatrix) | 2.51 GB | 24.27 | 31.69 |
| IQ2_XXS | unsloth UD-IQ2_XXS | 2.63 GB | 23.68 | 31.16 |
| IQ2_XXS | DASH-Q IQ2_XXS | 2.65 GB | 21.15 | 28.39 |
| IQ2_XS | llama.cpp IQ2_XS (imatrix) | 2.72 GB | 22.97 | 30.07 |
| IQ2_XS | DASH-Q IQ2_XS | 2.98 GB | 20.05 | 26.76 |
| IQ2_M | llama.cpp IQ2_M (imatrix) | 3.07 GB | 21.38 | 28.03 |
| IQ2_M | unsloth UD-IQ2_M | 3.13 GB | 21.10 | 27.99 |
| IQ2_M | DASH-Q IQ2_M | 3.14 GB | 19.72 | 26.48 |
| Q2_K_XL | llama.cpp Q2_K (imatrix) | 3.30 GB | 21.18 | 28.37 |
| Q2_K_XL | unsloth UD-Q2_K_XL | 3.54 GB | 20.49 | 26.98 |
| Q2_K_XL | DASH-Q Q2_K_XL | 3.38 GB | 19.63 | 26.24 |
llama-perplexity, context 2048; WikiText-2 test, C4 validation (256 x 2048 tokens).
Usage
llama-cli -m Seed-Coder-8B-Instruct-DASHQ-Q2_K_XL.gguf -ngl 99 -c 8192
License
Inherits the license of the base model (ByteDance-Seed/Seed-Coder-8B-Instruct).