Skip to content

Commit 36179ff

Browse files
mudlermudler
authored
fix(backend gallery): intel images for python-based backends, re-add exllama2 (#5928)
chore(backend gallery): fix intel images for python-based backends Signed-off-by: Ettore Di Giacinto <[email protected]>
1 parent d25145e commit 36179ff

File tree

2 files changed

+143
-238
lines changed

2 files changed

+143
-238
lines changed

‎.github/workflows/backend.yml‎

Lines changed: 77 additions & 100 deletions
Original file line numberDiff line numberDiff line change
@@ -381,24 +381,12 @@ jobs:
381381
dockerfile: "./backend/Dockerfile.python"
382382
context: "./backend"
383383
# sycl builds
384-
- build-type: 'sycl_f32'
385-
cuda-major-version: ""
386-
cuda-minor-version: ""
387-
platforms: 'linux/amd64'
388-
tag-latest: 'auto'
389-
tag-suffix: '-gpu-intel-sycl-f32-rerankers'
390-
runs-on: 'ubuntu-latest'
391-
base-image: "quay.io/go-skynet/intel-oneapi-base:latest"
392-
skip-drivers: 'false'
393-
backend: "rerankers"
394-
dockerfile: "./backend/Dockerfile.python"
395-
context: "./backend"
396-
- build-type: 'sycl_f16'
384+
- build-type: 'intel'
397385
cuda-major-version: ""
398386
cuda-minor-version: ""
399387
platforms: 'linux/amd64'
400388
tag-latest: 'auto'
401-
tag-suffix: '-gpu-intel-sycl-f16-rerankers'
389+
tag-suffix: '-gpu-intel-rerankers'
402390
runs-on: 'ubuntu-latest'
403391
base-image: "quay.io/go-skynet/intel-oneapi-base:latest"
404392
skip-drivers: 'false'
@@ -429,157 +417,85 @@ jobs:
429417
backend: "llama-cpp"
430418
dockerfile: "./backend/Dockerfile.llama-cpp"
431419
context: "./"
432-
- build-type: 'sycl_f32'
433-
cuda-major-version: ""
434-
cuda-minor-version: ""
435-
platforms: 'linux/amd64'
436-
tag-latest: 'auto'
437-
tag-suffix: '-gpu-intel-sycl-f32-vllm'
438-
runs-on: 'ubuntu-latest'
439-
base-image: "quay.io/go-skynet/intel-oneapi-base:latest"
440-
skip-drivers: 'false'
441-
backend: "vllm"
442-
dockerfile: "./backend/Dockerfile.python"
443-
context: "./backend"
444-
- build-type: 'sycl_f16'
420+
- build-type: 'intel'
445421
cuda-major-version: ""
446422
cuda-minor-version: ""
447423
platforms: 'linux/amd64'
448424
tag-latest: 'auto'
449-
tag-suffix: '-gpu-intel-sycl-f16-vllm'
425+
tag-suffix: '-gpu-intel-vllm'
450426
runs-on: 'ubuntu-latest'
451427
base-image: "quay.io/go-skynet/intel-oneapi-base:latest"
452428
skip-drivers: 'false'
453429
backend: "vllm"
454430
dockerfile: "./backend/Dockerfile.python"
455431
context: "./backend"
456-
- build-type: 'sycl_f32'
457-
cuda-major-version: ""
458-
cuda-minor-version: ""
459-
platforms: 'linux/amd64'
460-
tag-latest: 'auto'
461-
tag-suffix: '-gpu-intel-sycl-f32-transformers'
462-
runs-on: 'ubuntu-latest'
463-
base-image: "quay.io/go-skynet/intel-oneapi-base:latest"
464-
skip-drivers: 'false'
465-
backend: "transformers"
466-
dockerfile: "./backend/Dockerfile.python"
467-
context: "./backend"
468-
- build-type: 'sycl_f16'
432+
- build-type: 'intel'
469433
cuda-major-version: ""
470434
cuda-minor-version: ""
471435
platforms: 'linux/amd64'
472436
tag-latest: 'auto'
473-
tag-suffix: '-gpu-intel-sycl-f16-transformers'
437+
tag-suffix: '-gpu-intel-transformers'
474438
runs-on: 'ubuntu-latest'
475439
base-image: "quay.io/go-skynet/intel-oneapi-base:latest"
476440
skip-drivers: 'false'
477441
backend: "transformers"
478442
dockerfile: "./backend/Dockerfile.python"
479443
context: "./backend"
480-
- build-type: 'sycl_f32'
444+
- build-type: 'intel'
481445
cuda-major-version: ""
482446
cuda-minor-version: ""
483447
platforms: 'linux/amd64'
484448
tag-latest: 'auto'
485-
tag-suffix: '-gpu-intel-sycl-f32-diffusers'
449+
tag-suffix: '-gpu-intel-diffusers'
486450
runs-on: 'ubuntu-latest'
487451
base-image: "quay.io/go-skynet/intel-oneapi-base:latest"
488452
skip-drivers: 'false'
489453
backend: "diffusers"
490454
dockerfile: "./backend/Dockerfile.python"
491455
context: "./backend"
492456
# SYCL additional backends
493-
- build-type: 'sycl_f32'
494-
cuda-major-version: ""
495-
cuda-minor-version: ""
496-
platforms: 'linux/amd64'
497-
tag-latest: 'auto'
498-
tag-suffix: '-gpu-intel-sycl-f32-kokoro'
499-
runs-on: 'ubuntu-latest'
500-
base-image: "quay.io/go-skynet/intel-oneapi-base:latest"
501-
skip-drivers: 'false'
502-
backend: "kokoro"
503-
dockerfile: "./backend/Dockerfile.python"
504-
context: "./backend"
505-
- build-type: 'sycl_f16'
457+
- build-type: 'intel'
506458
cuda-major-version: ""
507459
cuda-minor-version: ""
508460
platforms: 'linux/amd64'
509461
tag-latest: 'auto'
510-
tag-suffix: '-gpu-intel-sycl-f16-kokoro'
462+
tag-suffix: '-gpu-intel-kokoro'
511463
runs-on: 'ubuntu-latest'
512464
base-image: "quay.io/go-skynet/intel-oneapi-base:latest"
513465
skip-drivers: 'false'
514466
backend: "kokoro"
515467
dockerfile: "./backend/Dockerfile.python"
516468
context: "./backend"
517-
- build-type: 'sycl_f32'
518-
cuda-major-version: ""
519-
cuda-minor-version: ""
520-
platforms: 'linux/amd64'
521-
tag-latest: 'auto'
522-
tag-suffix: '-gpu-intel-sycl-f32-faster-whisper'
523-
runs-on: 'ubuntu-latest'
524-
base-image: "quay.io/go-skynet/intel-oneapi-base:latest"
525-
skip-drivers: 'false'
526-
backend: "faster-whisper"
527-
dockerfile: "./backend/Dockerfile.python"
528-
context: "./backend"
529-
- build-type: 'sycl_f16'
469+
- build-type: 'intel'
530470
cuda-major-version: ""
531471
cuda-minor-version: ""
532472
platforms: 'linux/amd64'
533473
tag-latest: 'auto'
534-
tag-suffix: '-gpu-intel-sycl-f16-faster-whisper'
474+
tag-suffix: '-gpu-intel-faster-whisper'
535475
runs-on: 'ubuntu-latest'
536476
base-image: "quay.io/go-skynet/intel-oneapi-base:latest"
537477
skip-drivers: 'false'
538478
backend: "faster-whisper"
539479
dockerfile: "./backend/Dockerfile.python"
540480
context: "./backend"
541-
- build-type: 'sycl_f32'
542-
cuda-major-version: ""
543-
cuda-minor-version: ""
544-
platforms: 'linux/amd64'
545-
tag-latest: 'auto'
546-
tag-suffix: '-gpu-intel-sycl-f32-coqui'
547-
runs-on: 'ubuntu-latest'
548-
base-image: "quay.io/go-skynet/intel-oneapi-base:latest"
549-
skip-drivers: 'false'
550-
backend: "coqui"
551-
dockerfile: "./backend/Dockerfile.python"
552-
context: "./backend"
553-
- build-type: 'sycl_f16'
481+
- build-type: 'intel'
554482
cuda-major-version: ""
555483
cuda-minor-version: ""
556484
platforms: 'linux/amd64'
557485
tag-latest: 'auto'
558-
tag-suffix: '-gpu-intel-sycl-f16-coqui'
486+
tag-suffix: '-gpu-intel-coqui'
559487
runs-on: 'ubuntu-latest'
560488
base-image: "quay.io/go-skynet/intel-oneapi-base:latest"
561489
skip-drivers: 'false'
562490
backend: "coqui"
563491
dockerfile: "./backend/Dockerfile.python"
564492
context: "./backend"
565-
- build-type: 'sycl_f32'
566-
cuda-major-version: ""
567-
cuda-minor-version: ""
568-
platforms: 'linux/amd64'
569-
tag-latest: 'auto'
570-
tag-suffix: '-gpu-intel-sycl-f32-bark'
571-
runs-on: 'ubuntu-latest'
572-
base-image: "quay.io/go-skynet/intel-oneapi-base:latest"
573-
skip-drivers: 'false'
574-
backend: "bark"
575-
dockerfile: "./backend/Dockerfile.python"
576-
context: "./backend"
577-
- build-type: 'sycl_f16'
493+
- build-type: 'intel'
578494
cuda-major-version: ""
579495
cuda-minor-version: ""
580496
platforms: 'linux/amd64'
581497
tag-latest: 'auto'
582-
tag-suffix: '-gpu-intel-sycl-f16-bark'
498+
tag-suffix: '-gpu-intel-bark'
583499
runs-on: 'ubuntu-latest'
584500
base-image: "quay.io/go-skynet/intel-oneapi-base:latest"
585501
skip-drivers: 'false'
@@ -930,6 +846,67 @@ jobs:
930846
backend: "rfdetr"
931847
dockerfile: "./backend/Dockerfile.python"
932848
context: "./backend"
849+
# exllama2
850+
- build-type: ''
851+
cuda-major-version: ""
852+
cuda-minor-version: ""
853+
platforms: 'linux/amd64'
854+
tag-latest: 'auto'
855+
tag-suffix: '-cpu-exllama2'
856+
runs-on: 'ubuntu-latest'
857+
base-image: "ubuntu:22.04"
858+
skip-drivers: 'false'
859+
backend: "exllama2"
860+
dockerfile: "./backend/Dockerfile.python"
861+
context: "./backend"
862+
- build-type: 'cublas'
863+
cuda-major-version: "12"
864+
cuda-minor-version: "0"
865+
platforms: 'linux/amd64'
866+
tag-latest: 'auto'
867+
tag-suffix: '-gpu-nvidia-cuda-12-exllama2'
868+
runs-on: 'ubuntu-latest'
869+
base-image: "ubuntu:22.04"
870+
skip-drivers: 'false'
871+
backend: "exllama2"
872+
dockerfile: "./backend/Dockerfile.python"
873+
context: "./backend"
874+
- build-type: 'cublas'
875+
cuda-major-version: "11"
876+
cuda-minor-version: "7"
877+
platforms: 'linux/amd64'
878+
tag-latest: 'auto'
879+
tag-suffix: '-gpu-nvidia-cuda-11-exllama2'
880+
runs-on: 'ubuntu-latest'
881+
base-image: "ubuntu:22.04"
882+
skip-drivers: 'false'
883+
backend: "exllama2"
884+
dockerfile: "./backend/Dockerfile.python"
885+
context: "./backend"
886+
- build-type: 'intel'
887+
cuda-major-version: ""
888+
cuda-minor-version: ""
889+
platforms: 'linux/amd64'
890+
tag-latest: 'auto'
891+
tag-suffix: '-gpu-intel-exllama2'
892+
runs-on: 'ubuntu-latest'
893+
base-image: "quay.io/go-skynet/intel-oneapi-base:latest"
894+
skip-drivers: 'false'
895+
backend: "exllama2"
896+
dockerfile: "./backend/Dockerfile.python"
897+
context: "./backend"
898+
- build-type: 'hipblas'
899+
cuda-major-version: ""
900+
cuda-minor-version: ""
901+
platforms: 'linux/amd64'
902+
skip-drivers: 'true'
903+
tag-latest: 'auto'
904+
tag-suffix: '-gpu-hipblas-exllama2'
905+
base-image: "rocm/dev-ubuntu-22.04:6.1"
906+
runs-on: 'ubuntu-latest'
907+
backend: "exllama2"
908+
dockerfile: "./backend/Dockerfile.python"
909+
context: "./backend"
933910
# runs out of space on the runner
934911
# - build-type: 'hipblas'
935912
# cuda-major-version: ""

0 commit comments

Comments
 (0)