Lumpiasty/llama.cpp
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

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

Browse Source
This commit is contained in:
Masashi Yoshimura
2026-06-30 17:20:04 +09:00
committed by GitHub
parent 86b94708f2
commit 6c5de1cc83
6 changed files with 135 additions and 3 deletions
Download Patch File Download Diff File Expand all files Collapse all files
ggml/src/ggml-webgpu/ggml-webgpu.cpp
+3
View File
@@ -4056,6 +4056,7 @@ static bool ggml_webgpu_supported_qtype(ggml_type type) {
case GGML_TYPE_IQ4_NL:
case GGML_TYPE_IQ4_XS:
case GGML_TYPE_MXFP4:
case GGML_TYPE_NVFP4:
return true;
default:
return false;
@@ -4156,6 +4157,7 @@ static bool ggml_backend_webgpu_device_supports_op(ggml_backend_dev_t dev, const
case GGML_TYPE_IQ4_NL:
case GGML_TYPE_IQ4_XS:
case GGML_TYPE_MXFP4:
case GGML_TYPE_NVFP4:
supports_op = true;
break;
default:
@@ -4196,6 +4198,7 @@ static bool ggml_backend_webgpu_device_supports_op(ggml_backend_dev_t dev, const
case GGML_TYPE_IQ4_NL:
case GGML_TYPE_IQ4_XS:
case GGML_TYPE_MXFP4:
case GGML_TYPE_NVFP4:
supports_op = true;
break;
default: