Skip to content

Actions: shalinib-ibm/llama.cpp

Actions

EditorConfig Checker

Actions

Loading...
Loading

Show workflow options

Create status badge

Loading
editorconfig.yml
41 workflow runs
41 workflow runs

Filter by Event

Filter by Status

Filter by Branch

Filter by Actor

Q4/Q8 Tiled Gemm Optimization. EditorConfig Checker #41: Pull request #22 synchronize by shalinib-ibm
20s q4_q8_llama_opt
q4_q8_llama_opt
20s
Q4/Q8 Tiled Gemm Optimization. EditorConfig Checker #40: Pull request #22 opened by shalinib-ibm
28s q4_q8_llama_opt
q4_q8_llama_opt
28s
Q4 Tiled Gemm Optimization EditorConfig Checker #39: Pull request #21 opened by shalinib-ibm
26s llama_q4_opt
llama_q4_opt
26s
ggml-cuda: use passed ops instead of hardcoded ops (#16712) EditorConfig Checker #38: Commit 061f0ef pushed by shalinib-ibm
27s master
master
27s
Prepack all k blocks of Matrices A and B EditorConfig Checker #37: Pull request #20 opened by shalinib-ibm
13m 3s q4_gemm_prepack_kblocks
q4_gemm_prepack_kblocks
13m 3s
Pre Compute scale factor once and reuse inside kenrel. EditorConfig Checker #36: Pull request #19 opened by shalinib-ibm
21s q4_gemm_precompute_scale
q4_gemm_precompute_scale
21s
Implement a 16x8 kernel for Q4 GEMM EditorConfig Checker #35: Pull request #18 opened by shalinib-ibm
18s q4_gemm_trail3
q4_gemm_trail3
18s
Q4 tiled Gemm Implementation EditorConfig Checker #34: Pull request #17 opened by shalinib-ibm
18s q4_gemm_trail2
q4_gemm_trail2
18s
Q4 gemm trail1 EditorConfig Checker #33: Pull request #16 synchronize by shalinib-ibm
24s q4_gemm_trail1
q4_gemm_trail1
24s
Q4 gemm trail1 EditorConfig Checker #32: Pull request #16 opened by shalinib-ibm
16s q4_gemm_trail1
q4_gemm_trail1
16s
PowerPC: Sgemm Optimization EditorConfig Checker #31: Pull request #15 synchronize by shalinib-ibm
20s tb_ppc_sgemm_opt
tb_ppc_sgemm_opt
20s
PowerPC: Sgemm Optimization EditorConfig Checker #30: Pull request #15 synchronize by shalinib-ibm
21s tb_ppc_sgemm_opt
tb_ppc_sgemm_opt
21s
PowerPC: Sgemm Optimization EditorConfig Checker #29: Pull request #15 synchronize by shalinib-ibm
22s tb_ppc_sgemm_opt
tb_ppc_sgemm_opt
22s
PowerPC: Sgemm Optimization EditorConfig Checker #28: Pull request #15 synchronize by shalinib-ibm
31s tb_ppc_sgemm_opt
tb_ppc_sgemm_opt
31s
PowerPC: Sgemm Optimization EditorConfig Checker #27: Pull request #15 synchronize by shalinib-ibm
27s tb_ppc_sgemm_opt
tb_ppc_sgemm_opt
27s
PowerPC: Sgemm Optimization EditorConfig Checker #26: Pull request #15 opened by shalinib-ibm
18s tb_ppc_sgemm_opt
tb_ppc_sgemm_opt
18s
context : fix index overflow on huge outputs (#15080) EditorConfig Checker #25: Commit ee3a9fc pushed by shalinib-ibm
19s master
master
19s
POC:Avoid PackTranspose TinyBLAS_PPC in MMA kernel EditorConfig Checker #24: Pull request #14 opened by shalinib-ibm
19s poc_fp32_no_trans_A
poc_fp32_no_trans_A
19s
PowerPC: Replace static_assert with runtime assertion EditorConfig Checker #23: Pull request #13 opened by shalinib-ibm
9m 33s gcc-11-workaround
gcc-11-workaround
9m 33s
llama : add jinja template for rwkv-world (#14665) EditorConfig Checker #22: Commit 0d92267 pushed by shalinib-ibm
19s master
master
19s
Exp: Perf Benefit with xvi8ger4pp signed version EditorConfig Checker #21: Pull request #12 opened by shalinib-ibm
20s xvi8ger4pp_signed_signed
xvi8ger4pp_signed_signed
20s
Move packing from inside MMA kernels EditorConfig Checker #20: Pull request #11 opened by shalinib-ibm
1m 36s ppc_f32_move_pack
ppc_f32_move_pack
1m 36s
Refactor: Move matrix packing outside GEMM kernels EditorConfig Checker #19: Pull request #10 opened by shalinib-ibm
17s prepack_tinyblas_ppc
prepack_tinyblas_ppc
17s
Avoid PackTranspose calls Fp32 tiny blas kernels EditorConfig Checker #18: Pull request #9 opened by shalinib-ibm
20s tinyblas_ppc_opt
tinyblas_ppc_opt
20s
Optimize ggml_vec_dot_q8_q8 for ppc EditorConfig Checker #17: Pull request #8 opened by shalinib-ibm
20s splat_rem_vec_dot_q8_q8
splat_rem_vec_dot_q8_q8
20s