aboutsummaryrefslogtreecommitdiff
path: root/qwen_asr_kernels.c (follow)
Commit message (Collapse)AuthorAgeFilesLines
* kernels: drop asr_ prefix namingOnur Gündüz2026-06-141-1745/+0
|
* kernels: fix OpenBLAS thread contention and numerical instabilityOnur Gündüz2026-06-141-2/+30
|
* kernels: parallelize large rmsnorm rowsOnur Gündüz2026-06-141-5/+48
|
* late: add packed maxsim scoringOnur Gündüz2026-06-141-4/+8
|
* kernels: vectorize bf16 widening on armOnur Gündüz2026-06-141-0/+14
|
* kernels: use sgemv for single-row f32 linearOnur Gündüz2026-06-141-4/+10
|
* kernels: add neon rope apply pathOnur Gündüz2026-06-141-0/+21
|
* cpu: reuse tiled attention scratch buffersOnur Gündüz2026-06-141-12/+46
|
* cpu: tile packed attention through blasOnur Gündüz2026-06-141-54/+110
|
* kernels,cpu: fuse tiny BF16 gate up projectionsOnur Gündüz2026-06-141-0/+93
|
* kernels,cpu: fuse small BF16 QKV projectionsOnur Gündüz2026-06-141-15/+40
|
* kernels: validate real BF16 artifactsOnur Gündüz2026-06-141-0/+52
|
* cpu: fuse CPU embedding hot pathsOnur Gündüz2026-06-141-0/+26
|
* kernels: avoid tiny packed attention dispatchOnur Gündüz2026-06-141-1/+7
|
* all: support batched embedding attentionOnur Gündüz2026-06-141-0/+95
|
* Refactor and enhance pplx_embed and pplx_embed_mlx for improved model supportOnur Gündüz2026-03-281-5/+0
| | | | | | | | | | | | - Updated `pplx_embed.h` to support new embedding models: `pplx-embed-v1-4b` and `pplx-embed-context-v1-4b`. - Simplified model configuration structure and constants for better maintainability. - Enhanced comments for clarity on model architecture and configuration. - Modified `pplx_embed_mlx.c` to streamline loading of model weights and configuration from `config.json`. - Improved memory management and error handling in the MLX backend. - Updated embedding computation to support dynamic hidden sizes based on model configuration. - Added `pplx_mlx_config` function to retrieve model configuration details. - Updated README.md - Added .gitignore
* InitOnur Gündüz2026-03-281-0/+1247