Wagner Bruna 25fab4113e refactor: handle GGML_VK_VISIBLE_DEVICES at the Python level (#2179) ... All C++ handling code currently: - build a comma-separated list from the info_vulkan array - if GGML_VK_VISIBLE_DEVICES isn't set - set GGML_VK_VISIBLE_DEVICES to the list Once set, GGML_VK_VISIBLE_DEVICES affects the whole process. So this can be done in the same way at the Python level, before all loading functions. Caveat: load_model had the default `inputs.vulkan_info = "0"`, so the default GPU would be "0" only when loading a text model.		2026-05-02 23:10:29 +08:00
..
acestep	refactor: handle GGML_VK_VISIBLE_DEVICES at the Python level (#2179)	2026-05-02 23:10:29 +08:00
qwen3tts	use original precision for q3tts	2026-04-30 17:28:11 +08:00
sdcpp	refactor: handle GGML_VK_VISIBLE_DEVICES at the Python level (#2179)	2026-05-02 23:10:29 +08:00
tools
ttscpp	hack to allow kokoro to remain functional even with much higher GGML_SCHED_MAX_SPLIT_INPUTS	2026-04-19 20:40:07 +08:00
whispercpp	refactor: handle GGML_VK_VISIBLE_DEVICES at the Python level (#2179)	2026-05-02 23:10:29 +08:00
embeddings_adapter.cpp	refactor: handle GGML_VK_VISIBLE_DEVICES at the Python level (#2179)	2026-05-02 23:10:29 +08:00
ggml_v1.c	OpenBSD also needs alloca.h (#2012)	2026-03-05 12:32:31 +08:00
ggml_v1.h
ggml_v2-cuda-legacy.cu	try fix compile issues rocm (+4 squashed commit)	2025-09-21 22:31:09 +08:00
ggml_v2-cuda-legacy.h
ggml_v2-cuda.cu	try fix compile issues rocm (+4 squashed commit)	2025-09-21 22:31:09 +08:00
ggml_v2-cuda.h
ggml_v2.c	remove clblast, part 2	2026-01-23 14:09:46 +08:00
ggml_v2.h
ggml_v3-cuda.cu	try fix compile issues rocm (+4 squashed commit)	2025-09-21 22:31:09 +08:00
ggml_v3-cuda.h
ggml_v3.c
ggml_v3.h
gpt2_v1.cpp
gpt2_v2.cpp	Adjust int types in format strings (#2009)	2026-03-06 19:06:18 +08:00
gpt2_v3.cpp
gptj_v1.cpp
gptj_v2.cpp	Adjust int types in format strings (#2009)	2026-03-06 19:06:18 +08:00
gptj_v3.cpp	Adjust int types in format strings (#2009)	2026-03-06 19:06:18 +08:00
llama-util.h
llama_v2-util.h
llama_v2.cpp
llama_v2.h
llama_v3.cpp
llama_v3.h
mpt_v3.cpp
neox_v2.cpp
neox_v3.cpp
otherarch.h	added preliminary support for reasoning budget	2026-04-18 11:56:33 +08:00
rwkv_v2.cpp	Adjust int types in format strings (#2009)	2026-03-06 19:06:18 +08:00
rwkv_v2.h
rwkv_v3.cpp
rwkv_v3.h
rwkv_vocab.cpp	move embeds into their own dir, detach sd vocab into separate files	2025-10-03 14:21:09 +08:00
tts_adapter.cpp	refactor: handle GGML_VK_VISIBLE_DEVICES at the Python level (#2179)	2026-05-02 23:10:29 +08:00
utils.cpp	handle SWA conflicting with rewind, increased default SWA padding.	2026-04-16 17:00:26 +08:00
utils.h	handle SWA conflicting with rewind, increased default SWA padding.	2026-04-16 17:00:26 +08:00