ggml WebGPU: add support for quantization types #15440
reeselevine
merged
25 commits into
ggml-org:master
from
reeselevine:master
Begin work on set_rows
6a6135cc
Work on set rows
b2dbfcdc
Add error buffers for reporting unsupported SET_ROWS indices
248f7a51
Remove extra comments
4ad09861
Work on templating for different types in shaders
6355137c
Work on shader type generation
831ea3c3
Working q4_0 mul_mat and some templating for different types
688b51db
Add q4_0_f16 matmul and fix device init
1aa40f1a
Add matmul support for basic quantization types
c3611f9d
Add q2_k and q3_k quantization
de4da871
Add rest of k-quants
d76e562b
Get firt i-quant working
e2380e25
Closer to supporting all i-quants
2a3b9ee2
Support rest of i-quants
57c26b17
Merge remote-tracking branch 'origin/master' into types
7a2ae489
Cleanup code
51252f02
Merge pull request #2 from reeselevine/types
985508e5
Fix python formatting
6552e2e4
debug
65bebd31
Bugfix for memset
16df269a
Add padding to end of buffers on creation
10babfd7
Simplify bit-shifting
7a323b08
Merge pull request #3 from reeselevine/fixes
d1b0ffe9
Update usage of StringView
d6903035
ggerganov
approved these changes
on 2025-08-22
Merge remote-tracking branch 'upstream/master'
1fcc4047
Login to write a write a comment.
Login via GitHub