hybrid-llama/turboquant/conversion
Marvin 1dd0700988 hybrid-llama: merge ik_llama IQK CPU GEMM into TurboQuant fork
Base: AtomicBot-ai/atomic-llama-cpp-turboquant @ cd5609390. IQK source: ikawrakow/ik_llama.cpp @ fe215a8c (ggml/src/iqk only).

- GGML_IQK_MUL_MAT / GGML_IQK_FLASH_ATTENTION options (default OFF)

- 57 IQK repacked types, blocks, traits; IQK hooks in ggml_compute_forward_mul_mat

- ggml-cpu with IQK ON builds and links; IQK OFF build unaffected

Assisted-by: opencode (Muse Spark)
2026-09-05 18:09:49 -03:00
..
__init__.py hybrid-llama: merge ik_llama IQK CPU GEMM into TurboQuant fork 2026-09-05 18:09:49 -03:00
afmoe.py hybrid-llama: merge ik_llama IQK CPU GEMM into TurboQuant fork 2026-09-05 18:09:49 -03:00
arctic.py hybrid-llama: merge ik_llama IQK CPU GEMM into TurboQuant fork 2026-09-05 18:09:49 -03:00
baichuan.py hybrid-llama: merge ik_llama IQK CPU GEMM into TurboQuant fork 2026-09-05 18:09:49 -03:00
bailingmoe.py hybrid-llama: merge ik_llama IQK CPU GEMM into TurboQuant fork 2026-09-05 18:09:49 -03:00
base.py hybrid-llama: merge ik_llama IQK CPU GEMM into TurboQuant fork 2026-09-05 18:09:49 -03:00
bert.py hybrid-llama: merge ik_llama IQK CPU GEMM into TurboQuant fork 2026-09-05 18:09:49 -03:00
bitnet.py hybrid-llama: merge ik_llama IQK CPU GEMM into TurboQuant fork 2026-09-05 18:09:49 -03:00
bloom.py hybrid-llama: merge ik_llama IQK CPU GEMM into TurboQuant fork 2026-09-05 18:09:49 -03:00
chameleon.py hybrid-llama: merge ik_llama IQK CPU GEMM into TurboQuant fork 2026-09-05 18:09:49 -03:00
chatglm.py hybrid-llama: merge ik_llama IQK CPU GEMM into TurboQuant fork 2026-09-05 18:09:49 -03:00
codeshell.py hybrid-llama: merge ik_llama IQK CPU GEMM into TurboQuant fork 2026-09-05 18:09:49 -03:00
cogvlm.py hybrid-llama: merge ik_llama IQK CPU GEMM into TurboQuant fork 2026-09-05 18:09:49 -03:00
command_r.py hybrid-llama: merge ik_llama IQK CPU GEMM into TurboQuant fork 2026-09-05 18:09:49 -03:00
dbrx.py hybrid-llama: merge ik_llama IQK CPU GEMM into TurboQuant fork 2026-09-05 18:09:49 -03:00
deci.py hybrid-llama: merge ik_llama IQK CPU GEMM into TurboQuant fork 2026-09-05 18:09:49 -03:00
deepseek.py hybrid-llama: merge ik_llama IQK CPU GEMM into TurboQuant fork 2026-09-05 18:09:49 -03:00
dots1.py hybrid-llama: merge ik_llama IQK CPU GEMM into TurboQuant fork 2026-09-05 18:09:49 -03:00
dotsocr.py hybrid-llama: merge ik_llama IQK CPU GEMM into TurboQuant fork 2026-09-05 18:09:49 -03:00
dream.py hybrid-llama: merge ik_llama IQK CPU GEMM into TurboQuant fork 2026-09-05 18:09:49 -03:00
ernie.py hybrid-llama: merge ik_llama IQK CPU GEMM into TurboQuant fork 2026-09-05 18:09:49 -03:00
exaone.py hybrid-llama: merge ik_llama IQK CPU GEMM into TurboQuant fork 2026-09-05 18:09:49 -03:00
falcon.py hybrid-llama: merge ik_llama IQK CPU GEMM into TurboQuant fork 2026-09-05 18:09:49 -03:00
falcon_h1.py hybrid-llama: merge ik_llama IQK CPU GEMM into TurboQuant fork 2026-09-05 18:09:49 -03:00
gemma.py hybrid-llama: merge ik_llama IQK CPU GEMM into TurboQuant fork 2026-09-05 18:09:49 -03:00
glm.py hybrid-llama: merge ik_llama IQK CPU GEMM into TurboQuant fork 2026-09-05 18:09:49 -03:00
gpt2.py hybrid-llama: merge ik_llama IQK CPU GEMM into TurboQuant fork 2026-09-05 18:09:49 -03:00
gpt_oss.py hybrid-llama: merge ik_llama IQK CPU GEMM into TurboQuant fork 2026-09-05 18:09:49 -03:00
gptneox.py hybrid-llama: merge ik_llama IQK CPU GEMM into TurboQuant fork 2026-09-05 18:09:49 -03:00
granite.py hybrid-llama: merge ik_llama IQK CPU GEMM into TurboQuant fork 2026-09-05 18:09:49 -03:00
grok.py hybrid-llama: merge ik_llama IQK CPU GEMM into TurboQuant fork 2026-09-05 18:09:49 -03:00
grovemoe.py hybrid-llama: merge ik_llama IQK CPU GEMM into TurboQuant fork 2026-09-05 18:09:49 -03:00
hunyuan.py hybrid-llama: merge ik_llama IQK CPU GEMM into TurboQuant fork 2026-09-05 18:09:49 -03:00
inkling.py hybrid-llama: merge ik_llama IQK CPU GEMM into TurboQuant fork 2026-09-05 18:09:49 -03:00
internlm.py hybrid-llama: merge ik_llama IQK CPU GEMM into TurboQuant fork 2026-09-05 18:09:49 -03:00
internvl.py hybrid-llama: merge ik_llama IQK CPU GEMM into TurboQuant fork 2026-09-05 18:09:49 -03:00
jais.py hybrid-llama: merge ik_llama IQK CPU GEMM into TurboQuant fork 2026-09-05 18:09:49 -03:00
jamba.py hybrid-llama: merge ik_llama IQK CPU GEMM into TurboQuant fork 2026-09-05 18:09:49 -03:00
januspro.py hybrid-llama: merge ik_llama IQK CPU GEMM into TurboQuant fork 2026-09-05 18:09:49 -03:00
kimi_k3.py hybrid-llama: merge ik_llama IQK CPU GEMM into TurboQuant fork 2026-09-05 18:09:49 -03:00
kimi_linear.py hybrid-llama: merge ik_llama IQK CPU GEMM into TurboQuant fork 2026-09-05 18:09:49 -03:00
kimivl.py hybrid-llama: merge ik_llama IQK CPU GEMM into TurboQuant fork 2026-09-05 18:09:49 -03:00
laguna.py hybrid-llama: merge ik_llama IQK CPU GEMM into TurboQuant fork 2026-09-05 18:09:49 -03:00
lfm2.py hybrid-llama: merge ik_llama IQK CPU GEMM into TurboQuant fork 2026-09-05 18:09:49 -03:00
lighton_ocr.py hybrid-llama: merge ik_llama IQK CPU GEMM into TurboQuant fork 2026-09-05 18:09:49 -03:00
llada.py hybrid-llama: merge ik_llama IQK CPU GEMM into TurboQuant fork 2026-09-05 18:09:49 -03:00
llama.py hybrid-llama: merge ik_llama IQK CPU GEMM into TurboQuant fork 2026-09-05 18:09:49 -03:00
llama4.py hybrid-llama: merge ik_llama IQK CPU GEMM into TurboQuant fork 2026-09-05 18:09:49 -03:00
llava.py hybrid-llama: merge ik_llama IQK CPU GEMM into TurboQuant fork 2026-09-05 18:09:49 -03:00
maincoder.py hybrid-llama: merge ik_llama IQK CPU GEMM into TurboQuant fork 2026-09-05 18:09:49 -03:00
mamba.py hybrid-llama: merge ik_llama IQK CPU GEMM into TurboQuant fork 2026-09-05 18:09:49 -03:00
mellum.py hybrid-llama: merge ik_llama IQK CPU GEMM into TurboQuant fork 2026-09-05 18:09:49 -03:00
mimo.py hybrid-llama: merge ik_llama IQK CPU GEMM into TurboQuant fork 2026-09-05 18:09:49 -03:00
minicpm.py hybrid-llama: merge ik_llama IQK CPU GEMM into TurboQuant fork 2026-09-05 18:09:49 -03:00
minimax.py hybrid-llama: merge ik_llama IQK CPU GEMM into TurboQuant fork 2026-09-05 18:09:49 -03:00
mistral.py hybrid-llama: merge ik_llama IQK CPU GEMM into TurboQuant fork 2026-09-05 18:09:49 -03:00
mistral3.py hybrid-llama: merge ik_llama IQK CPU GEMM into TurboQuant fork 2026-09-05 18:09:49 -03:00
mpt.py hybrid-llama: merge ik_llama IQK CPU GEMM into TurboQuant fork 2026-09-05 18:09:49 -03:00
nanbeige.py hybrid-llama: merge ik_llama IQK CPU GEMM into TurboQuant fork 2026-09-05 18:09:49 -03:00
nemotron.py hybrid-llama: merge ik_llama IQK CPU GEMM into TurboQuant fork 2026-09-05 18:09:49 -03:00
olmo.py hybrid-llama: merge ik_llama IQK CPU GEMM into TurboQuant fork 2026-09-05 18:09:49 -03:00
openelm.py hybrid-llama: merge ik_llama IQK CPU GEMM into TurboQuant fork 2026-09-05 18:09:49 -03:00
orion.py hybrid-llama: merge ik_llama IQK CPU GEMM into TurboQuant fork 2026-09-05 18:09:49 -03:00
pangu.py hybrid-llama: merge ik_llama IQK CPU GEMM into TurboQuant fork 2026-09-05 18:09:49 -03:00
phi.py hybrid-llama: merge ik_llama IQK CPU GEMM into TurboQuant fork 2026-09-05 18:09:49 -03:00
pixtral.py hybrid-llama: merge ik_llama IQK CPU GEMM into TurboQuant fork 2026-09-05 18:09:49 -03:00
plamo.py hybrid-llama: merge ik_llama IQK CPU GEMM into TurboQuant fork 2026-09-05 18:09:49 -03:00
plm.py hybrid-llama: merge ik_llama IQK CPU GEMM into TurboQuant fork 2026-09-05 18:09:49 -03:00
qwen.py hybrid-llama: merge ik_llama IQK CPU GEMM into TurboQuant fork 2026-09-05 18:09:49 -03:00
qwen3vl.py hybrid-llama: merge ik_llama IQK CPU GEMM into TurboQuant fork 2026-09-05 18:09:49 -03:00
qwenvl.py hybrid-llama: merge ik_llama IQK CPU GEMM into TurboQuant fork 2026-09-05 18:09:49 -03:00
refact.py hybrid-llama: merge ik_llama IQK CPU GEMM into TurboQuant fork 2026-09-05 18:09:49 -03:00
rwkv.py hybrid-llama: merge ik_llama IQK CPU GEMM into TurboQuant fork 2026-09-05 18:09:49 -03:00
sarashina2.py hybrid-llama: merge ik_llama IQK CPU GEMM into TurboQuant fork 2026-09-05 18:09:49 -03:00
smallthinker.py hybrid-llama: merge ik_llama IQK CPU GEMM into TurboQuant fork 2026-09-05 18:09:49 -03:00
smolvlm.py hybrid-llama: merge ik_llama IQK CPU GEMM into TurboQuant fork 2026-09-05 18:09:49 -03:00
stablelm.py hybrid-llama: merge ik_llama IQK CPU GEMM into TurboQuant fork 2026-09-05 18:09:49 -03:00
starcoder.py hybrid-llama: merge ik_llama IQK CPU GEMM into TurboQuant fork 2026-09-05 18:09:49 -03:00
step3.py hybrid-llama: merge ik_llama IQK CPU GEMM into TurboQuant fork 2026-09-05 18:09:49 -03:00
t5.py hybrid-llama: merge ik_llama IQK CPU GEMM into TurboQuant fork 2026-09-05 18:09:49 -03:00
talkie.py hybrid-llama: merge ik_llama IQK CPU GEMM into TurboQuant fork 2026-09-05 18:09:49 -03:00
ultravox.py hybrid-llama: merge ik_llama IQK CPU GEMM into TurboQuant fork 2026-09-05 18:09:49 -03:00
wavtokenizer.py hybrid-llama: merge ik_llama IQK CPU GEMM into TurboQuant fork 2026-09-05 18:09:49 -03:00
xverse.py hybrid-llama: merge ik_llama IQK CPU GEMM into TurboQuant fork 2026-09-05 18:09:49 -03:00
youtuvl.py hybrid-llama: merge ik_llama IQK CPU GEMM into TurboQuant fork 2026-09-05 18:09:49 -03:00