XPU: `from_pretrained(device_map=...)` fails under WSL2 because `caching_allocator_warmup` does not guard `mem_get_info`

Continuous batching auto-switches to flash_attention_3 on unsupported GPUs (Turing/T4), crashing — regression from 5.14.1

Potential host synchronization in deepspeed_sp_compute_loss due to Python conditional on CUDA tensor

CohereASR training-loss double-shift bug (same pattern as Moonshine #46784)

`apply_chat_template` crashes with `IndexError` when passed an empty list

[Gemma 4] Gemma4UnifiedForConditionalGeneration text-only inference produces degenerate output (token repetition collapse)

[Bug] Incorrect error message for T5 model family's decoder input validation

Regression: `convert_tokens_to_ids` is much slower in v5 than v4 for slow tokenizers with many added tokens

Transformers v5.9.0 breaks support for pytorch with USE_DISTRIBUTED=0

Qwen3.5-9B video input fails with apply_chat_template — Video features and video tokens do not match