ld_openelm.c

1.支持batch

2.q8量化

3.avx2和avx512加速

4.python wrapper

5.零依赖

6.单文件完成推理

7.cpu版本，开启-O3优化的情况下加速1.5倍

8.cpu版本，q8量化，开启-O3优化的情况下加速3倍

9.混合精度

Name		Name	Last commit message	Last commit date
Latest commit History 24 Commits
.gitignore		.gitignore
LICENSE		LICENSE
README.md		README.md
export_openelm.py		export_openelm.py
export_prompt.py		export_prompt.py
export_tokens.py		export_tokens.py
openelm.c		openelm.c
openelm.config		openelm.config
openelm.cu		openelm.cu
openelm.keys		openelm.keys
openelm.net		openelm.net
openelm_kv.c		openelm_kv.c
openelm_kv_avx2.c		openelm_kv_avx2.c
openelm_kv_avx512.c		openelm_kv_avx512.c
openelm_kv_q40.c		openelm_kv_q40.c
openelm_kv_q80.c		openelm_kv_q80.c
openelm_kv_q80_v1.c		openelm_kv_q80_v1.c
openelm_kv_q80_v2.c		openelm_kv_q80_v2.c
openelm_kv_v1.c		openelm_kv_v1.c
openelm_kv_v2.c		openelm_kv_v2.c
openelm_tokenizer.src		openelm_tokenizer.src
openelm_v1.cu		openelm_v1.cu
openelm_v2.cu		openelm_v2.cu
run.py		run.py
run_kv.py		run_kv.py
run_kv_q80.py		run_kv_q80.py

Provide feedback