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
..
_index.en.md	feat: docs revamp (#7313)	2025-11-19 22:21:20 +01:00
_index.md	docs(agents): adopt kernel's AI coding assistants policy	2026-04-19 22:50:54 +00:00
ai-coding-assistants.md	docs(agents): adopt kernel's AI coding assistants policy	2026-04-19 22:50:54 +00:00
api-errors.md	feat: add users and authentication support (#9061)	2026-03-19 21:40:51 +01:00
architecture.md	feat: docs revamp (#7313)	2025-11-19 22:21:20 +01:00
binaries.md	chore: drop bark which is unmaintained (#8207)	2026-01-25 09:26:40 +01:00
cli-reference.md	feat: add node reconciler, allow to schedule to group of nodes, min/max autoscaler (#9186)	2026-03-31 08:28:56 +02:00
compatibility-table.md	chore(whisperx): drop ROCm/hipblas build target (#9474)	2026-04-21 21:50:18 +02:00
nvidia-l4t.md	chore(docs): update docs with cuda 13 instructions and the new vibevoice backend	2025-12-25 10:00:07 +01:00
shell-completion.md	Update shell completion documentation URL	2026-03-08 09:33:36 +01:00
system-info.md	feat: Add documentation for undocumented API endpoints (#8852)	2026-03-08 17:59:33 +01:00