| Commit message (Expand) | Author | Age | Files | Lines |
| * | server: resolve contextual separator from vocab, test contextual path | Onur Gündüz | 2026-06-14 | 5 | -11/+154 |
| * | tests: drive the HTTP request path in-process in test_server | Onur Gündüz | 2026-06-14 | 4 | -103/+548 |
| * | tokenizer: ASCII class table for pretokenizer codepoint checks | Onur Gündüz | 2026-06-14 | 2 | -9/+51 |
| * | build: add make coverage (clang source-based, llvm-cov) | Onur Gündüz | 2026-06-14 | 2 | -16/+75 |
| * | build: rename make targets blas->cpu, mlx->metal, shared->lib | Onur Gündüz | 2026-06-14 | 3 | -26/+26 |
| * | server: default micro-batch cap to 32; add scheduler sweep driver | Onur Gündüz | 2026-06-14 | 1 | -3/+6 |
| * | server: do not idle-reap clients with requests in flight | Onur Gündüz | 2026-06-14 | 1 | -1/+4 |
| * | late: batch MaxSim as grouped GEMM plus row-max reduction | Onur Gündüz | 2026-06-14 | 1 | -2/+60 |
| * | cpu: move BF16 widening scratch into the workspace | Onur Gündüz | 2026-06-14 | 4 | -152/+61 |
| * | tests: hermetic BPE tokenizer test with fast/slow path equality | Onur Gündüz | 2026-06-14 | 3 | -0/+191 |
| * | tokenizer: integer-pair BPE merging with linked-list symbols | Onur Gündüz | 2026-06-14 | 2 | -5/+241 |
| * | tests: add server encoder unit tests via EMBED_SERVER_TEST include | Onur Gündüz | 2026-06-14 | 3 | -1/+97 |
| * | cuda: follow model file dtype for weight storage by default | Onur Gündüz | 2026-06-14 | 1 | -1/+15 |
| * | tests: make safetensors and bf16-model tests hermetic, join make test | Onur Gündüz | 2026-06-14 | 5 | -25/+273 |
| * | gitignore: bench binary | Onur Gündüz | 2026-06-14 | 1 | -0/+1 |
| * | tests,bench: move generated C checkers into standing files | Onur Gündüz | 2026-06-14 | 5 | -1/+448 |
| * | cli: update help text | Onur Gündüz | 2026-06-14 | 1 | -2/+0 |
| * | chore: rename sources; binaries pplx-embed, pplx-embed-server | Onur Gündüz | 2026-06-14 | 12 | -54/+386 |
| * | gitignore: track new binary and library names | Onur Gündüz | 2026-06-14 | 14 | -341/+4 |
| * | kernels: drop asr_ prefix naming | Onur Gündüz | 2026-06-14 | 19 | -57/+57 |
| * | tests: move kernel goldens into tests/, add make test | Onur Gündüz | 2026-06-14 | 1 | -0/+337 |
| * | cli: rename flags, add --stream, split serving into pplx-embed-server | Onur Gündüz | 2026-06-14 | 6 | -230/+441 |
| * | build: produce libpplxembed.a and a shared-library target | Onur Gündüz | 2026-06-14 | 1 | -5/+38 |
| * | cuda: store BF16 activations directly in norm and SiLU kernels | Onur Gündüz | 2026-06-14 | 1 | -51/+114 |
| * | build: allow FP reassociation/contraction, keep finite-math off | Onur Gündüz | 2026-06-14 | 1 | -3/+7 |
| * | server: fix unused-rc warning in plain BLAS builds | Onur Gündüz | 2026-06-14 | 1 | -1/+2 |
| * | build: add .clang-tidy configuration | Onur Gündüz | 2026-06-14 | 1 | -0/+13 |
| * | gitignore: add .vscode directory to ignore list | Onur Gündüz | 2026-06-14 | 1 | -0/+1 |
| * | gitignore: add private directory to ignore list | Onur Gündüz | 2026-06-14 | 1 | -0/+1 |
| * | mlx: mixed-precision BF16 inference (F32 residual, BF16 weights) | Onur Gündüz | 2026-06-14 | 1 | -7/+71 |
| * | cuda: add optional BF16 CUDA weight storage (--cuda-weights bf16) | Onur Gündüz | 2026-06-14 | 3 | -68/+162 |
| * | cuda: unify attention on one path; drop the max_seq>=64 dispatch | Onur Gündüz | 2026-06-14 | 1 | -73/+8 |
| * | cuda: add fast GEMM mode (--cuda-fast-gemm mode) | Onur Gündüz | 2026-06-14 | 3 | -17/+102 |
| * | cuda: clarify comment gemm_compute | Onur Gündüz | 2026-06-14 | 1 | -2/+3 |
| * | cuda: add opt-in BF16 tensor-core GEMM mode | Onur Gündüz | 2026-06-14 | 1 | -21/+34 |
| * | cuda: replace long-seq attention with cuBLAS GEMM attention | Onur Gündüz | 2026-06-14 | 1 | -158/+124 |
| * | cuda: replace tiled attention with warp-per-query reduction | Onur Gündüz | 2026-06-14 | 1 | -98/+105 |
| * | cuda: extend tiled attention to ragged micro-batches | Onur Gündüz | 2026-06-14 | 1 | -3/+105 |
| * | cuda: add tiled single attention | Onur Gündüz | 2026-06-14 | 1 | -4/+64 |
| * | cuda: optimize rmsnorm reduction | Onur Gündüz | 2026-06-14 | 1 | -34/+29 |
| * | cuda: add server contextual path | Onur Gündüz | 2026-06-14 | 5 | -21/+239 |
| * | cuda: optimize projection path | Onur Gündüz | 2026-06-14 | 1 | -87/+118 |
| * | cuda: fuse qkv projection | Onur Gündüz | 2026-06-14 | 1 | -44/+78 |
| * | cuda: accumulate residual projections | Onur Gündüz | 2026-06-14 | 1 | -21/+18 |
| * | cuda: optimize attention baseline | Onur Gündüz | 2026-06-14 | 1 | -44/+47 |
| * | cuda: add initial CUDA embedding backend | Onur Gündüz | 2026-06-14 | 6 | -68/+911 |
| * | kernels: fix OpenBLAS thread contention and numerical instability | Onur Gündüz | 2026-06-14 | 2 | -3/+33 |
| * | mlx: split ragged contextual batches | Onur Gündüz | 2026-06-14 | 1 | -3/+137 |
| * | mlx: tune ragged grouping for 4B | Onur Gündüz | 2026-06-14 | 1 | -5/+34 |
| * | mlx: prepare contiguous linear weights | Onur Gündüz | 2026-06-14 | 1 | -4/+21 |