Fresh
13Stale
0Expired
1Merged
6LFM2-VL: integration expectations after #50734 (A10, MI300X)
fix(qwen3_vl): size video budget for the padded frame count in smart_resize
[WIP] Add GS1 decision model support
Add DeepSeek-V4.1-Flash
[Chat Parsing] Quote bare keys that have whitespace around them with unquoted_keys
Migrate testers for 20 popular models to static small-config defaults
[Whisper] Fix EnglishTextNormalizer turning “1.5” into “one.5”
[SGLang CI] Fix the VLM teardown error on CPU (follow-up of #50913)
[SGLang CI] Deselect a GPU-only VLM test
[vLLM CI] Fix the FireRedASR2 and MTP test failures (missing dependency, GPU-only test)
Remove now useless keys_to_ignore_at_inference in all configs
Simplify cache handling in generate
Set _skip_keys_device_placement directly inside PreTrainedModel
Restore SentencePiece nmt_nfkc normalization in DebertaV2Tokenizer
Engine hooks for the quantization plan loader: load_serialized, sharding op, in-place params
Modular cleanup post merge
[serve] Forward reasoning history to chat templates in get_processor_inputs_from_messages
Add Hy4-flash (hy_v4_flash)
Fix flaky DeepseekV32 CUDA graph capture failure (GC inside the capture, torch 2.15)
[vLLM CI] Only give HF_TOKEN to the jobs that run the tests