diff options
| author | Kawrakow <48489457+ikawrakow@users.noreply.github.com> | 2024-02-26 18:28:38 +0200 |
|---|---|---|
| committer | GitHub <noreply@github.com> | 2024-02-26 18:28:38 +0200 |
| commit | a33e6a0d2a66104ea9a906bdbf8a94d050189d91 (patch) | |
| tree | 30478b4a0b1792d1af66c5d64e2c3c4fa1af74ab /ggml.h | |
| parent | 47bb7b48c7cec9d8f57d56812ce811ec130b89a3 (diff) | |
Adding IQ2_S and IQ2_M to complete coverage of the 2-3 bit quantization range (#5721)
* Adding IQ2_S and IQ2_M as a single cumulative commit
* Update examples/quantize/quantize.cpp
Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
---------
Co-authored-by: Iwan Kawrakow <iwan.kawrakow@gmail.com>
Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
Diffstat (limited to 'ggml.h')
| -rw-r--r-- | ggml.h | 2 |
1 files changed, 2 insertions, 0 deletions
@@ -351,6 +351,7 @@ extern "C" { GGML_TYPE_IQ1_S = 19, GGML_TYPE_IQ4_NL = 20, GGML_TYPE_IQ3_S = 21, + GGML_TYPE_IQ2_S = 22, GGML_TYPE_I8, GGML_TYPE_I16, GGML_TYPE_I32, @@ -391,6 +392,7 @@ extern "C" { GGML_FTYPE_MOSTLY_IQ1_S = 18, // except 1d tensors GGML_FTYPE_MOSTLY_IQ4_NL = 19, // except 1d tensors GGML_FTYPE_MOSTLY_IQ3_S = 20, // except 1d tensors + GGML_FTYPE_MOSTLY_IQ2_S = 21, // except 1d tensors }; // available tensor operations: |