vulkan: Add logging for bf16 features to ggml_vk_print_gpu_info (#13274) #14707
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
github-actions
bot
added
Vulkan
0cc4m
approved these changes
Jul 19, 2025
Collaborator
0cc4m
left a comment
0cc4m
left a comment
There was a problem hiding this comment.
The code looks fine to me. I don't think bf16 has major performance implications currently, unless maybe if it is used extensively in a model, but knowing whether it's supported natively can't hurt.
Contributor
|
nice! don't forget to add similar fp8 Vulkan support reporting once llama.cpp adds use of VK_EXT_shader_float8 to it's Vulkan backend! |
Collaborator
This is only relevant if the new datatype makes any difference in the backend. bf16 is being used in specific cases, float8 would need a shader to use it first. Edit: On second read, that is what you said. I misread it, sorry. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Issue #13274 requested logging whether the VK_KHR_bfloat16 is supported at runtime. I have added the detection for the feature to ggml_vk_print_gpu_info, and tested it on windows with the latest Nvidia GeForce Game Ready Driver (where it reported 0) as well as the Nvidia Vulkan 1.4 Developer Beta Driver (where it reported 1).
I have not tested Linux or any GPUs other than my laptop RTX 4060.