005907fd9e
bind mount source without "./"
2026-05-26 09:40:25 +02:00
0abe6a54dc
create missing intermediate cache dir
2026-05-26 09:38:32 +02:00
b0b7f85b0d
use bind mounts to keep gh actions cache within working dir
2026-05-26 09:36:05 +02:00
0786400490
"fix?" heredoc in 7.2.3 in buildah bud
2026-05-25 21:57:53 +02:00
065abcf7ef
no push
2026-05-25 21:53:32 +02:00
e0eb69c035
tar suid
2026-05-25 21:53:01 +02:00
e9cbeab90e
use correct secret names
2026-05-25 20:14:18 +02:00
1f7e13f574
checkout v6
2026-05-25 20:11:45 +02:00
805de308dc
test2
2026-05-25 20:10:45 +02:00
0e5dca4350
test
2026-05-25 19:43:24 +02:00
0b4ea35d26
Squashed commit of the following:
...
commit 9bb311392cjulian@wefers.page >
Date: Mon May 25 19:22:21 2026 +0200
attempt fix podman login
commit ef2a27498djulian@wefers.page >
Date: Mon May 25 19:11:51 2026 +0200
use github runner cache instead of caching to docker hub
commit aa9faa0573julian@wefers.page >
Date: Mon May 25 18:57:00 2026 +0200
6.4.4 refinements
commit 28b7f6b0dcjulian@wefers.page >
Date: Mon May 25 18:31:43 2026 +0200
cache dnf per variant to avoid concurrent invalidation
commit dee34a478ajulian@wefers.page >
Date: Mon May 25 18:15:49 2026 +0200
comments
commit aef3efb059julian@wefers.page >
Date: Mon May 25 18:15:45 2026 +0200
amdvlk add -j nproc
commit ed193f48c5julian@wefers.page >
Date: Mon May 25 18:01:19 2026 +0200
remove default ",rw" option for RUN mounts
commit bd606776d8julian@wefers.page >
Date: Mon May 25 17:59:49 2026 +0200
refinement for radv
commit e4c070fea4julian@wefers.page >
Date: Mon May 25 17:31:30 2026 +0200
remove more dnf cleans
commit 61db525030julian@wefers.page >
Date: Sat May 23 13:44:45 2026 +0200
comments, fix nightly, ld errors remain
commit 908880302ajulian@wefers.page >
Date: Sat May 23 12:44:30 2026 +0200
undo chunkah attempt
commit c465571533julian@wefers.page >
Date: Wed May 20 19:50:31 2026 +0200
work
commit 860e22f049julian@wefers.page >
Date: Tue May 19 20:50:33 2026 +0200
do rocm7-nightlies
commit 928a54a891julian@wefers.page >
Date: Mon May 18 17:38:52 2026 +0200
do vulkan-amdvlk
commit bb91b23915julian@wefers.page >
Date: Mon May 18 17:34:05 2026 +0200
do vulkan-radv, give id to the llama caches to avoid cross-backend pollution
commit 97f34d2989julian@wefers.page >
Date: Mon May 18 17:28:10 2026 +0200
add 6.4.4 dockerfile, revert unneeded changes in 7.2.3 file
commit b8a5752a46julian@wefers.page >
Date: Mon May 18 11:55:23 2026 +0200
fix run stage emptying out the dnf cache again
commit de42ea152ajulian@wefers.page >
Date: Sun May 17 18:47:30 2026 +0200
fix build cache in workflow
commit 88b30cc5fejulian@wefers.page >
Date: Sun May 17 18:44:51 2026 +0200
rem personal test build script
commit d2de478763julian@wefers.page >
Date: Sun May 17 18:37:30 2026 +0200
fix caching
commit eadf9039e8julian@wefers.page >
Date: Sun May 17 17:59:39 2026 +0200
add layers in build workload
commit aeb3da66a4julian@wefers.page >
Date: Sun May 17 17:57:14 2026 +0200
more caching
commit f057441277julian@wefers.page >
Date: Sun May 17 15:37:40 2026 +0200
changes
commit 49adce7e79julian@wefers.page >
Date: Sun May 17 14:43:27 2026 +0200
test
commit 7f94e6d6cbjulian@wefers.page >
Date: Sun May 17 14:36:08 2026 +0200
test
commit 83f0b884efjulian@wefers.page >
Date: Sun May 17 14:34:13 2026 +0200
install podman
commit b557f81157julian@wefers.page >
Date: Sun May 17 14:22:03 2026 +0200
use full image, on gitea normal image is missing podman, buildah
commit e1ef8f841fjulian@wefers.page >
Date: Sun May 17 14:12:43 2026 +0200
more caching
commit cf16448886donato.capitella@reversec.com >
Date: Fri May 22 11:14:47 2026 +0100
fix(ci): restore strict exit status checks (remove 139)
commit b53818bb2adonato.capitella@reversec.com >
Date: Fri May 22 11:12:50 2026 +0100
fix(ci): allow exit code 139 (SIGSEGV) in container smoke tests
commit b03855d91edonato.capitella@reversec.com >
Date: Fri May 22 10:41:09 2026 +0100
fix(ci): adjust container smoke test exit-code validation for CPU runners
commit 5f8f9ca40edonato.capitella@reversec.com >
Date: Fri May 22 09:40:48 2026 +0100
ci: add smoke tests for built images and ignore research directory
commit a4bcc396badonato.capitella@reversec.com >
Date: Fri May 22 08:58:39 2026 +0100
fix: update llama.cpp library installation path to /usr/local/lib64 and improve runtime linking across Dockerfiles
commit 08aa7bc026donato.capitella@reversec.com >
Date: Mon May 18 07:42:50 2026 +0100
refactor: prune obsolete benchmark result logs and update existing entries
2026-05-25 19:23:00 +02:00
bb2a64e3d0
use buildah
2026-05-25 19:19:06 +02:00
71abff5a0d
feat: add interactive granular benchmark details to UI and update MTP summary data format
2026-05-17 11:03:23 +01:00
2a4cb50b52
chore: update benchmark execution script parameters and paths
2026-05-17 09:16:03 +01:00
451723bab8
docs: update recommended boot parameters to use amd_iommu=off for improved performance
2026-05-15 19:08:21 +01:00
a1ef189df0
feat: add MTP benchmark page with dedicated data summary and visualization components
2026-05-15 11:51:02 +01:00
adf4d1a5fa
mtp-results
2026-05-15 10:56:57 +01:00
fa27bf663c
refactor: update display name logic in benchmark script to consistently use filenames and strip multi-part suffix
2026-05-15 10:21:24 +01:00
d1e5e0b3b0
fix: enable SO_REUSEADDR on socket to prevent address binding errors in benchmark utility
2026-05-15 09:22:06 +01:00
adf72bdda7
refactor: replace immediate port kill with a 15-second wait for port release in cleanup function
2026-05-15 08:58:16 +01:00
765f635381
feat: improve container exit error reporting by adding log retrieval and manual debug guidance
2026-05-15 08:48:39 +01:00
b1d526b775
feat: add SIGINT handler for container cleanup and enable replace flag for podman execution
2026-05-15 08:43:34 +01:00
c0e33115ca
feat: implement container status monitoring and enhanced log reporting in health check loop
2026-05-15 08:38:24 +01:00
7e6c30dae6
feat: add pre-flight cleanup and port conflict resolution to benchmark runner
2026-05-15 08:36:12 +01:00
8bc5b8c5e5
feat: add automated MTP benchmark runner for llama-server via podman containers
2026-05-15 08:07:10 +01:00
dc948d766a
feat: add benchmark script for MTP acceptance and throughput testing
2026-05-14 21:02:05 +01:00
3e3f3674a8
feat: add MTP (Multi-Token Prediction) support via new ROCm 7.2.3 and Vulkan RADV toolboxes
2026-05-14 20:09:03 +01:00
7320eb3f00
updated 7.2.3 benchs
2026-05-12 12:38:43 +01:00
8648f93ad3
updated benchs
2026-05-12 12:32:07 +01:00
ed889def9d
Support Debian in toolbox/distrobox command check ( #88 )
...
* Support Debian in toolbox command check
Debian requires distrobox, like Ubuntu. This is a simple change to add support for it.
* Consolidation of logic
copilot complained, so used this as an opportunity to collapse unnecessary evals.
2026-05-12 12:25:59 +01:00
2e3dc657d2
chore: update ROCm version to 7.2.3 and remove deprecated pr21344 toolbox
2026-05-11 19:40:30 +01:00
0f9c2c85be
Update issue templates
2026-05-06 18:06:18 +01:00
07d2131d8c
added @64k benchmarks
2026-05-03 16:20:42 +01:00
1bffd6505f
feat: add longctx65536 support to standard and RPC benchmark scripts
2026-05-01 20:19:02 +01:00
d20bb42b04
updated results
2026-04-29 06:45:26 +01:00
73be068e85
feat: upgrade ROCm toolboxes to 7.2.2 and update documentation and CI configurations
2026-04-26 16:25:44 +01:00
1421e87060
feat: add Plausible analytics script to documentation index page
2026-04-21 09:47:16 +01:00
9016c0f8f8
update benchs
2026-04-15 16:54:34 +01:00
66a3314c22
refactor: update MODEL_DIR path to use absolute home directory reference
2026-04-15 11:39:35 +01:00
9707a15df7
feat: add benchmark results for rocm-7_2_1-pr21344 and update results metadata
2026-04-15 11:39:10 +01:00
c2754a810a
fix vulkan builds: add spirv-headers-devel to builder deps
2026-04-15 09:33:13 +01:00
2c2c36d3da
add rocm-7.2.1-pr21344 toolbox (gfx1151 MMQ/MMVQ tile + nwarp tuning)
...
Adds a new toolbox variant based on PR #21344 (pedapudi/llama.cpp@gfx1151-opt)
which tunes MMQ tile sizes (x_max=48, y=64) and warp counts (nwarps=4) for
RDNA3_5 gfx1151, yielding up to +100% prefill throughput at small batch sizes.
Also adds BMI2/FMA/F16C CPU SIMD flags and GGML_CUDA_FA_ALL_QUANTS=ON to match
the benchmark build used in the PR. Wire up CI (build matrix + prune), the
refresh script, and run_benchmarks.sh so results land alongside rocm-7.2.1.
2026-04-15 09:23:58 +01:00
14fae26ad0
add minimax m2.7 benchmarks
2026-04-15 08:09:12 +01:00
4b3c02a405
Update README.md
2026-04-12 21:33:46 +01:00
d74db71362
archvied old multi-node benchmarks
2026-04-11 11:20:30 +01:00
7aa6e6dea9
update benchmarks
2026-04-11 11:18:45 +01:00
a821bcb91d
chore: update rocm-7.2 benchmark configuration to version 7.2.1
2026-04-10 11:48:27 +01:00
c129a04a1c
refactor: remove hblt0 benchmark support and associated comparison scripts
2026-04-10 11:23:06 +01:00
5acf54cd67
fix: Update HuggingFace download commands ( #61 )
2026-04-10 10:56:00 +01:00
1dea385f6a
fix: remove trailing backtick causing syntax error in prune-old-toolboxes workflow
2026-04-09 19:00:44 +01:00
Julian Wefers