Eugene Rakhmatulin
|
7d232a305a
|
Reverted to Torch 2.9.1 in the source build to address #24
|
2026-01-30 10:43:12 -08:00 |
|
Eugene Rakhmatulin
|
458439706a
|
Build flashinfer from source
|
2026-01-30 09:05:22 -08:00 |
|
Eugene Rakhmatulin
|
ef0f996df6
|
Bumped base image version; reverted Triton to 3.5.1
|
2026-01-29 23:14:43 -08:00 |
|
Eugene Rakhmatulin
|
0ac438b4dd
|
Some optimizations
|
2026-01-29 22:08:05 -08:00 |
|
Eugene Rakhmatulin
|
46fecd172a
|
added missing dependancy
|
2026-01-29 17:01:17 -08:00 |
|
Eugene Rakhmatulin
|
159460af0c
|
Migrated dockerfiles to pytorch-base image
|
2026-01-29 15:47:07 -08:00 |
|
Eugene Rakhmatulin
|
e817f3dbec
|
Updated Triton version to 3.6.0
|
2026-01-26 14:24:58 -08:00 |
|
Eugene Rakhmatulin
|
25a16ef6c2
|
Fixed #11 and #12 - added a new dependency for OpenCV
|
2026-01-19 12:07:15 -08:00 |
|
Eugene Rakhmatulin
|
1139a37324
|
Added transformers v5 support
|
2025-12-21 22:41:03 -08:00 |
|
Eugene Rakhmatulin
|
11db634aad
|
Switch to uv in the main Dockerfile
|
2025-12-21 13:28:40 -08:00 |
|
Eugene Rakhmatulin
|
dfe426e912
|
Add support for pre-release FlashInfer packages in Docker builds
|
2025-12-20 23:13:26 -08:00 |
|
Eugene Rakhmatulin
|
a83200573a
|
Enhance Dockerfile: limit ccache size, enable compression, and optimize git repo size
|
2025-12-20 15:29:37 -08:00 |
|
Eugene Rakhmatulin
|
fbb1bf73d5
|
Switching to flashinfer 0.6.x pre-release wheels
|
2025-12-20 13:28:06 -08:00 |
|
Christopher Owen
|
a13a9f6806
|
Limit build parallelism to reduce OOM situations
|
2025-12-18 13:36:35 +01:00 |
|
TeskaLabs Admin
|
f1abfb85b6
|
Bump of the version
|
2025-12-16 17:58:48 +00:00 |
|
Eugene Rakhmatulin
|
0606b1b984
|
Refactor Triton and vLLM reference handling in Dockerfile and build script
|
2025-12-14 23:28:08 -08:00 |
|
eugr
|
4551795908
|
Fixed missing Infiniband dependency, added CuDNN
|
2025-12-14 21:49:50 -08:00 |
|
eugr
|
33720fc9d6
|
Use no-build-isolation for Triton Kernels build
|
2025-12-14 18:35:26 -08:00 |
|
eugr
|
dc614dc6ae
|
Separated Triton build into a dedicated phase for better caching
|
2025-12-14 10:32:28 -08:00 |
|
eugr
|
25f759fec8
|
Optimized triton caching
|
2025-12-14 09:26:10 -08:00 |
|
eugr
|
e8a12da072
|
Build triton from source; add TRITON_SHA argument to specify triton release, and add timing statistics
|
2025-12-14 00:30:50 -08:00 |
|
eugr
|
a8217a1fd8
|
Improved dependency handling
|
2025-12-13 22:41:30 -08:00 |
|
eugr
|
cc3e73feb1
|
Improved caching
|
2025-12-13 21:34:57 -08:00 |
|
eugr
|
76a8e92c86
|
Multistage build with caching
|
2025-12-13 21:18:26 -08:00 |
|
eugr
|
37c12cf9e4
|
Removed MiniMax M2 patch since the fix is merged into main
|
2025-12-11 13:23:30 -08:00 |
|
eugr
|
5fba205db4
|
Implemented a temporary patch for recently broken MiniMax-M2 (in builds after 12/10) for some quants.
|
2025-12-11 11:13:05 -08:00 |
|
eugr
|
b10ed739fe
|
formatting changes
|
2025-11-29 10:04:12 -08:00 |
|
eugr
|
6a66a4b66f
|
Added patch to allow fastsafetensors in cluster config
|
2025-11-26 21:25:04 -08:00 |
|
eugr
|
549214e6ed
|
Added missing Infiniband and RDMA libraries
|
2025-11-25 16:14:08 -08:00 |
|
eugr
|
a96a3a2dac
|
Removed temporary patch for NVFP4 quants support as it's been merged into main
|
2025-11-25 12:48:58 -08:00 |
|
eugr
|
4c976375c5
|
Added missing dependencies; added dashboard support for Ray clusters
|
2025-11-24 21:13:06 -08:00 |
|
eugr
|
399948a725
|
Added missing modules for flashinfer
|
2025-11-24 17:02:04 -08:00 |
|
eugr
|
d3fd2e69fd
|
Updated Dockerfile with additional deps
|
2025-11-24 15:47:20 -08:00 |
|
eugr
|
f5141974ae
|
Fixed cluster script and small fix for Dockerfilewq
|
2025-11-24 15:45:04 -08:00 |
|
eugr
|
3ecca4d2b7
|
Updated Dockerfile to include 2 levels of cache busters, added the cluster script and README.
|
2025-11-24 15:21:08 -08:00 |
|
eugr
|
0ad880e0fe
|
Added clustering script
|
2025-11-24 11:53:38 -08:00 |
|
eugr
|
4e95bf6fa6
|
Initial commit
|
2025-11-24 11:19:37 -08:00 |
|