mirror of
https://github.com/ikawrakow/ik_llama.cpp.git
synced 2026-03-07 04:20:03 +00:00
154e0d75fc
* Merging mainline - WIP * Merging mainline - WIP AVX2 and CUDA appear to work. CUDA performance seems slightly (~1-2%) lower as it is so often the case with llama.cpp/ggml after some "improvements" have been made. * Merging mainline - fix Metal * Remove check --------- Co-authored-by: Iwan Kawrakow <iwan.kawrakow@gmail.com>
13 lines
428 B
Plaintext
13 lines
428 B
Plaintext
-r ../examples/llava/requirements.txt
|
|
-r ../examples/server/bench/requirements.txt
|
|
-r ../examples/server/tests/requirements.txt
|
|
|
|
-r ./requirements-compare-llama-bench.txt
|
|
-r ./requirements-pydantic.txt
|
|
-r ./requirements-test-tokenizer-random.txt
|
|
|
|
-r ./requirements-convert_hf_to_gguf.txt
|
|
-r ./requirements-convert_hf_to_gguf_update.txt
|
|
-r ./requirements-convert_legacy_llama.txt
|
|
-r ./requirements-convert_llama_ggml_to_gguf.txt
|