Ettore Di Giacinto 39573ecd2a chore(whisperx): drop ROCm/hipblas build target (#9474) ... whisperx has no upstream AMD GPU support and its core transcription path (faster-whisper -> ctranslate2) falls back to CPU on AMD since the PyPI ctranslate2 is CUDA-only. The torch rocm wheels would accelerate only the alignment/diarization stages, producing a misleadingly half-working image. Drop the hipblas variant rather than shipping a partially accelerated build users can't distinguish from the real thing. AMD hosts now fall through the capability map to cpu-whisperx / cpu-whisperx-development. Also removes the now-dangling rocm-whisperx assertion from pkg/system/capabilities_test.go and the ROCm mention from the whisperx row in docs/content/reference/compatibility-table.md. Assisted-by: Claude Code:claude-opus-4-7		2026-04-21 21:50:18 +02:00
..
capabilities.go	fix: try to add whisperx and faster-whisper for more variants (#9278)	2026-04-08 21:23:38 +02:00
capabilities_test.go	chore(whisperx): drop ROCm/hipblas build target (#9474)	2026-04-21 21:50:18 +02:00
state.go	feat(cli): add configurable backend image fallback tags via CLI options (#8817)	2026-03-08 21:16:37 +01:00
system_suite_test.go	fix: gate CUDA directory checks on GPU vendor to prevent false CUDA detection (#8942)	2026-03-12 07:53:39 +01:00