diff options
| author | Georgi Gerganov <ggerganov@gmail.com> | 2023-04-29 18:43:28 +0300 |
|---|---|---|
| committer | GitHub <noreply@github.com> | 2023-04-29 18:43:28 +0300 |
| commit | 214b6a35702a489e3738acd81fad6d46182d3036 (patch) | |
| tree | dac39b6d4bb7eaf958735a0dfb5ccabcbbb0821c /examples/quantize | |
| parent | 305eb5afd51325e3142c01c17431febb7c67de87 (diff) | |
ggml : adjust mul_mat_f16 work memory (#1226)
* llama : minor - remove explicity int64_t cast
* ggml : reduce memory buffer for F16 mul_mat when not using cuBLAS
* ggml : add asserts to guard for incorrect wsize
Diffstat (limited to 'examples/quantize')
0 files changed, 0 insertions, 0 deletions