Masashi Yoshimura 8f4646a63e ggml-webgpu: improve flash_attn_vec for quantized KV at long contexts (#25956) ...

* improve fa of quantized kv cache

* Fix some bugs and some comments.

* fix v type check and some comments

* Fix build error caused by rebasing

* editorconfig checking pass

2026-07-31 09:08:40 +03:00

..

add_id.wgsl

ggml-webgpu: Enables running gpt-oss-20b (#22906)

2026-05-12 07:27:40 -07:00

argmax.wgsl

ggml webgpu: support for backend sampling (#18880)

2026-01-16 16:12:43 -08:00

argsort_merge.wgsl

ggml webgpu: support for backend sampling (#18880)

2026-01-16 16:12:43 -08:00

argsort.wgsl

ggml webgpu: support for backend sampling (#18880)

2026-01-16 16:12:43 -08:00

binary.wgsl

[ggml-webgpu] Implement 2D workgroups for scale, binary, and unary ops (#24044)

2026-06-08 08:07:15 -07:00

common_decls.tmpl

ggml-webgpu: improve flash_attn_vec for quantized KV at long contexts (#25956)

2026-07-31 09:08:40 +03:00

concat.wgsl

[ggml-webgpu] Handle buffer overlap / buffer aliasing for concat operator (#24000)

2026-06-08 08:07:31 -07:00

conv2d_dw.wgsl

webgpu : add CONV_2D_DW (depthwise conv2d) kernel (#25847)

2026-07-22 17:24:44 +09:00

conv2d.wgsl

ggml-webgpu(shader): support conv2d kernels. (#21964)

2026-04-21 20:18:57 -07:00

cpy.wgsl

revert to using global_invocation_id for cpy shader (#23955)

2026-06-01 16:59:06 -07:00

cumsum.wgsl

ggml webgpu: support for backend sampling (#18880)

2026-01-16 16:12:43 -08:00

embed_wgsl.py

ggml-webgpu: port all AOT operators to JIT (#20728)

2026-03-31 15:38:16 -07:00

flash_attn_quant_staging.tmpl

ggml-webgpu: improve flash_attn_vec for quantized KV at long contexts (#25956)

2026-07-31 09:08:40 +03:00

flash_attn_tile.wgsl

ggml-webgpu: improve flash_attn_vec for quantized KV at long contexts (#25956)

2026-07-31 09:08:40 +03:00

flash_attn_vec_blk.wgsl

ggml-webgpu: enable FLASH_ATTN_EXT on browser without subgroup matrix (#22199)

2026-04-24 10:39:09 -07:00

flash_attn_vec_reduce.wgsl

ggml-webgpu: address precision issues for multimodal (#22808)

2026-05-12 07:27:04 -07:00

flash_attn_vec_split.wgsl

ggml-webgpu: improve flash_attn_vec for quantized KV at long contexts (#25956)

2026-07-31 09:08:40 +03:00

flash_attn.wgsl

ggml-webgpu: improve flash_attn_vec for quantized KV at long contexts (#25956)

2026-07-31 09:08:40 +03:00

gated_delta_net.wgsl

Remove padding and multiple D2D copies for MTP (#24086)

2026-06-10 23:21:16 +05:30

get_rows.wgsl

ggml-webgpu: add support for NVFP4 (#25143)

2026-06-30 17:20:04 +09:00

glu.wgsl

ggml-webgpu: Fix some binding alias issues to support all archs, fix recurrent-state-rollback test (#25931)

2026-07-28 21:13:06 +03:00

im2col.wgsl

ggml-webgpu: add support for im2col (#22259)

2026-04-22 20:17:41 -07:00

memset.wgsl

ggml WebGPU: add support for quantization types (#15440)

2025-08-22 11:28:03 -07:00

mul_mat_decls.tmpl

ggml-webgpu: add support for NVFP4 (#25143)

2026-06-30 17:20:04 +09:00

mul_mat_id_gather.wgsl

ggml-webgpu: Fix how to dispatch WG to some ops (#23750)

2026-05-27 09:48:12 -07:00

mul_mat_id_vec.wgsl

ggml-webgpu: improve MTP inference by using mat-vec path for small batches (#24811)

2026-06-23 17:13:55 +09:00

mul_mat_id.wgsl

ggml-webgpu: Fix dequantization helpers to not pass in pointers (#21872)

2026-04-15 09:14:40 -07:00

mul_mat_reg_tile.wgsl

ggml-webgpu: Fix dequantization helpers to not pass in pointers (#21872)

2026-04-15 09:14:40 -07:00

mul_mat_subgroup_matrix.wgsl

ggml-webgpu: Fix dequantization helpers to not pass in pointers (#21872)

2026-04-15 09:14:40 -07:00

mul_mat_vec_acc.tmpl

ggml-webgpu: add support for NVFP4 (#25143)

2026-06-30 17:20:04 +09:00

mul_mat_vec_q_acc.tmpl

ggml-webgpu: improve MTP inference by using mat-vec path for small batches (#24811)

2026-06-23 17:13:55 +09:00

mul_mat_vec.wgsl

ggml-webgpu: improve MTP inference by using mat-vec path for small batches (#24811)

2026-06-23 17:13:55 +09:00

pad.wgsl

ggml webgpu: support for backend sampling (#18880)

2026-01-16 16:12:43 -08:00

quant_inner_loops.tmpl

ggml-webgpu: FlashAttention refactor + standardize quantization support (#23834)

2026-06-04 08:05:04 +03:00

quantize_q8.wgsl

ggml-webgpu: improve MTP inference by using mat-vec path for small batches (#24811)

2026-06-23 17:13:55 +09:00

repeat.wgsl

ggml-webgpu: Add supports for GGML_OP_REPEAT (#20230)

2026-03-11 14:40:36 -07:00

rms_norm_mul.wgsl

ggml-webgpu: fix buffer aliasing for ssm_scan and refactor aliasing logic (#22456)

2026-04-28 07:27:17 -07:00

rope.wgsl

ggml-webgpu: port all AOT operators to JIT (#20728)

2026-03-31 15:38:16 -07:00

row_norm.wgsl

ggml-webgpu: add layer norm ops (#22406)

2026-05-03 20:52:53 -07:00

scale.wgsl

[ggml-webgpu] Implement 2D workgroups for scale, binary, and unary ops (#24044)

2026-06-08 08:07:15 -07:00

set_rows_quant.wgsl

ggml-webgpu: add q4_0/q8_0 SET_ROWS (#23760)

2026-05-29 14:14:11 -07:00

set_rows.wgsl

ggml-webgpu: add q4_0/q8_0 SET_ROWS (#23760)

2026-05-29 14:14:11 -07:00

set.wgsl

ggml webgpu: ops support for qwen3.5 (SET, TRI_SOLVE, SSM_CONV, GATED_DELTA_NET) + GET_ROWS optimization (#20687)

2026-03-19 08:45:28 -07:00

soft_max.wgsl

ggml-webgpu: port all AOT operators to JIT (#20728)

2026-03-31 15:38:16 -07:00

solve_tri.wgsl

ggml webgpu: ops support for qwen3.5 (SET, TRI_SOLVE, SSM_CONV, GATED_DELTA_NET) + GET_ROWS optimization (#20687)

2026-03-19 08:45:28 -07:00

ssm_conv.wgsl

ggml webgpu: ops support for qwen3.5 (SET, TRI_SOLVE, SSM_CONV, GATED_DELTA_NET) + GET_ROWS optimization (#20687)

2026-03-19 08:45:28 -07:00

ssm_scan.wgsl

ggml-webgpu: Fix some binding alias issues to support all archs, fix recurrent-state-rollback test (#25931)

2026-07-28 21:13:06 +03:00

sum_rows.wgsl

ggml webgpu: support for backend sampling (#18880)

2026-01-16 16:12:43 -08:00

unary.wgsl

[ggml-webgpu] Implement 2D workgroups for scale, binary, and unary ops (#24044)

2026-06-08 08:07:15 -07:00

upscale.wgsl

ggml-webgpu: add the upscale shader (#22419)

2026-04-30 22:22:18 -07:00