Skip to content

Actions: agray3/llama.cpp

flake8 Lint

Actions

Loading...
Loading

Show workflow options

Create status badge

Loading
36 workflow runs
36 workflow runs

Filter by Event

Filter by Status

Filter by Branch

Filter by Actor

Avoid rebuild of GGML graph for each token
flake8 Lint #10: Commit 78deb50 pushed by agray3
July 8, 2024 09:31 24s ag_ggml_graph_caching
July 8, 2024 09:31 24s
common : preallocate sampling token data vector (#8363)
flake8 Lint #9: Commit 470939d pushed by agray3
July 8, 2024 09:01 25s master
July 8, 2024 09:01 25s
Allow number of nodes in CUDA graph to change
flake8 Lint #8: Commit a2fb00d pushed by agray3
June 4, 2024 12:31 12m 17s ag_fix_numnodes_cuda_graph
June 4, 2024 12:31 12m 17s
Improve hipBLAS support in CMake (#7696)
flake8 Lint #7: Commit 987d743 pushed by agray3
June 4, 2024 12:24 28s master
June 4, 2024 12:24 28s
llama : avoid double token-to-piece cache (#7654)
flake8 Lint #6: Commit 549279d pushed by agray3
June 3, 2024 07:18 25s master
June 3, 2024 07:18 25s
ggml : fix loongarch build (O2 issue) (#7636)
flake8 Lint #5: Commit d5c0582 pushed by agray3
May 30, 2024 10:15 26s master
May 30, 2024 10:15 26s
metal : add GGML_OP_REPEAT kernels (#7557)
flake8 Lint #3: Commit 1d8fca7 pushed by agray3
May 27, 2024 11:54 3m 35s master
May 27, 2024 11:54 3m 35s
add build shared lib in win release package (#7438)
flake8 Lint #1: Commit 0df0aa8 pushed by agray3
May 24, 2024 11:13 44m 37s master
May 24, 2024 11:13 44m 37s