Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Adding bf16 support to CUDA #40

Merged
merged 3 commits into from
Sep 14, 2024
Merged

Adding bf16 support to CUDA #40

merged 3 commits into from
Sep 14, 2024

Conversation

ikawrakow
Copy link
Owner

@ikawrakow ikawrakow commented Sep 5, 2024

Haha, llama.cpp seems to not support bf16 on CUDA?

This PR adds it. It works fine on my RTX-4080, but I have no idea if it will work on older GPUs (if I understood correctly it should, with reduced performance), ROCm, etc.

Performance is the same as f16 for TG (TG-128 = 41.2 t/s for LLaMA-3.1-8B for both).

PP is lower but quite decent for prompt processing (PP-512(bf16) = 5250 t/s vs PP-512(f16) = 7250 t/s for LLaMA-3.1-8B). In any case, much better than running on the CPU for bf16 models.

@ikawrakow ikawrakow merged commit 6f11c95 into main Sep 14, 2024
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants