Skip to content

Actions: PrismML-Eng/llama.cpp

Actions

Check Pre-Tokenizer Hashes

Actions

Loading...
Loading

Show workflow options

Create status badge

Loading
pre-tokenizer-hashes.yml
14 workflow runs
14 workflow runs

Filter by Event

⌥ + click/return to exclude

Filter by Status

⌥ + click/return to exclude

Filter by Branch

⌥ + click/return to exclude

Filter by Actor

⌥ + click/return to exclude
server: Expose media_tag on /props endpoint. (#22028) Check Pre-Tokenizer Hashes #14: Commit 9e5647a pushed by khosravipasha
1m 11s master
master
1m 11s
Merge branch 'pr-21636' into prism-new Check Pre-Tokenizer Hashes #13: Commit 6df15a6 pushed by khosravipasha
55s prism
prism
55s
docs: listing qwen3-asr and qwen3-omni as supported (#21857) Check Pre-Tokenizer Hashes #12: Commit e974923 pushed by khosravipasha
1m 10s master
master
1m 10s
convert : force f16 or f32 on step3-vl conv weights (#21646) Check Pre-Tokenizer Hashes #11: Commit 1e9d771 pushed by khosravipasha
2m 4s master
master
2m 4s
CUDA: skip compilation of superfluous FA kernels (#21768) Check Pre-Tokenizer Hashes #10: Commit ff5ef82 pushed by khosravipasha
1m 0s master
master
1m 0s
vulkan: Support Q1_0 (#21539) Check Pre-Tokenizer Hashes #9: Commit 7b69125 pushed by khosravipasha
1m 21s master
master
1m 21s
CUDA: also store node->src->data ptrs for equality check (#21635) Check Pre-Tokenizer Hashes #8: Commit d12cc3d pushed by khosravipasha
1m 15s master
master
1m 15s
ggml: add Q1_0 1-bit quantization support (CPU) (#21273) Check Pre-Tokenizer Hashes #7: Commit 2e1f0a8 pushed by khosravipasha
1m 46s master
master
1m 46s
initial Q1_0 Metal backend Check Pre-Tokenizer Hashes #6: Pull request #14 synchronize by khosravipasha
1m 6s q1-metal
q1-metal
1m 6s
tuning q1_0 metal kernels Check Pre-Tokenizer Hashes #5: Commit 0776cd2 pushed by khosravipasha
1m 1s q1-metal
q1-metal
1m 1s
opencl: add q4_K gemm and gemv kernels for Adreno (#20919) Check Pre-Tokenizer Hashes #4: Commit 08f2145 pushed by khosravipasha
2m 26s master
master
2m 26s
models : move the token embedding norms to the first layer (#20943) Check Pre-Tokenizer Hashes #3: Commit 9f102a1 pushed by khosravipasha
52s master
master
52s
Merge pull request #1 from PrismML-Eng/mmq Check Pre-Tokenizer Hashes #2: Commit bc8122e pushed by khosravipasha
1m 26s v0.0.2-prism
v0.0.2-prism
1m 26s
CANN: support flash attention for head dim not multiple of 16, fix AL… Check Pre-Tokenizer Hashes #1: Commit 07ba6d2 pushed by khosravipasha
1m 12s master
master
1m 12s