Skip to content

RDNA4 updates #7

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.

Sign up for GitHub

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

Closed
pds-amd wants to merge 2 commits into from
Closed

RDNA4 updates #7

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
2914551
build: update rdna4 dockerfile
pds-amd Jul 21, 2026
14f0d9d
feat: update benchmark configurations for various models to enhance p…
pds-amd Jul 21, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
38 changes: 17 additions & 21 deletions benchmark_configs/xdit/flux.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -228,10 +228,10 @@
warmup_calls: 2
num_iterations: 5
guidance_scale: 0.0
attention_backend: aiter_flydsl
attention_backend: aiter_flydsl_fp8
use_torch_compile: true
use_fp8_gemms: true

cache_method: dbcache
- name: flux.usp_1k.2gpu.rdna4
tags: [rdna4, gfx1201]
model: black-forest-labs/FLUX.1-dev
Expand All @@ -246,14 +246,12 @@
warmup_calls: 2
num_iterations: 5
guidance_scale: 0.0
attention_backend: aiter_flydsl
attention_backend: aiter_flydsl_fp8
ulysses_degree: 2
ring_degree: 1
use_torch_compile: true
use_fp8_gemms: true
enable_tiling: true
use_parallel_vae: true

cache_method: dbcache
- name: flux.usp_1k.4gpu.rdna4
tags: [rdna4, gfx1201]
model: black-forest-labs/FLUX.1-dev
Expand All @@ -268,13 +266,12 @@
warmup_calls: 2
num_iterations: 5
guidance_scale: 0.0
attention_backend: aiter_flydsl
attention_backend: aiter_flydsl_fp8
ulysses_degree: 4
ring_degree: 1
use_torch_compile: true
use_fp8_gemms: true
enable_tiling: true
use_parallel_vae: true

cache_method: dbcache
- name: flux.2k.1gpu.rdna4
tags: [rdna4, gfx1201]
model: black-forest-labs/FLUX.1-dev
Expand All @@ -289,11 +286,11 @@
warmup_calls: 2
num_iterations: 5
guidance_scale: 0.0
attention_backend: aiter_flydsl
attention_backend: aiter_flydsl_fp8
use_torch_compile: true
enable_tiling: true
use_fp8_gemms: true

enable_tiling: true
cache_method: dbcache
- name: flux.usp_2k.2gpu.rdna4
tags: [rdna4, gfx1201]
model: black-forest-labs/FLUX.1-dev
Expand All @@ -308,14 +305,13 @@
warmup_calls: 2
num_iterations: 5
guidance_scale: 0.0
attention_backend: aiter_flydsl
attention_backend: aiter_flydsl_fp8
ulysses_degree: 2
ring_degree: 1
use_torch_compile: true
enable_tiling: true
use_parallel_vae: true
use_fp8_gemms: true

use_parallel_vae: true
enable_tiling: true
cache_method: dbcache
- name: flux.usp_2k.4gpu.rdna4
tags: [rdna4, gfx1201]
model: black-forest-labs/FLUX.1-dev
Expand All @@ -330,10 +326,10 @@
warmup_calls: 2
num_iterations: 5
guidance_scale: 0.0
attention_backend: aiter_flydsl
attention_backend: aiter_flydsl_fp8
ulysses_degree: 4
ring_degree: 1
use_torch_compile: true
use_fp8_gemms: true
enable_tiling: true
use_parallel_vae: true
enable_tiling: true
cache_method: dbcache
Loading