diff options
| author | Kawrakow <iwankawrakow@gmail.com> | 2024-09-29 09:03:52 +0300 |
|---|---|---|
| committer | GitHub <noreply@github.com> | 2024-09-29 09:03:52 +0300 |
| commit | fd20638bbcb4b1ba69783312bb78545fa418d3f2 (patch) | |
| tree | b9ff7dee2adeb73da4a77a7b55573efa7ccae931 /src | |
| parent | 1b789c983ac34679de67174662b25fb1227cebf2 (diff) | |
Allow bf16 kv-cache (#69)
On the CPU I get the exact same PPL with and without FA
using bf16 for kv-cache. But on CUDA the bf16 kv-cache
result is about the same as the fp16 kv-cache CPU result,
so I'm missing some conversion somewhere.
Co-authored-by: Iwan Kawrakow <iwan.kawrakow@gmail.com>
Diffstat (limited to 'src')
| -rw-r--r-- | src/llama.cpp | 2 |
1 files changed, 1 insertions, 1 deletions
diff --git a/src/llama.cpp b/src/llama.cpp index 94a939d8..dca03ade 100644 --- a/src/llama.cpp +++ b/src/llama.cpp @@ -16816,7 +16816,7 @@ struct llama_context * llama_new_context_with_model( params.flash_attn = false; } - if (params.type_v != GGML_TYPE_F16 && !params.flash_attn) { + if (params.type_v != GGML_TYPE_F16 && params.type_v != GGML_TYPE_BF16 && !params.flash_attn) { LLAMA_LOG_ERROR("%s: V cache quantization requires flash_attn\n", __func__); return nullptr; } |