Tal NirandClaude Opus 4.6 f75b61a9e9 [Voxtral Realtime] Fix engine crash on empty multimodal embeddings (#34862) ...

Signed-off-by: Tal Nir <tal@nervexneurotech.com>
Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>

2026-02-18 23:21:47 -08:00

..

instrumentator

[Kernel] Triton-based Top-k and Top-p sampler kernels (#33538)

2026-02-17 23:14:30 +00:00

llm

[Renderer] Move InputPreprocessor into Renderer (2/2) (#34560)

2026-02-17 05:29:01 -08:00

offline_mode

Fix offline test for Transformers v5 (#33682)

2026-02-03 12:07:24 +00:00

openai

[Voxtral Realtime] Fix engine crash on empty multimodal embeddings (#34862)

2026-02-18 23:21:47 -08:00

pooling

Extend ColBERT support to non-standard BERT backbones (#34170)

2026-02-13 09:53:09 +00:00

rpc

…

sagemaker

…

weight_transfer

[BUG] Fixing Weight Sync unit test (#34841)

2026-02-18 17:20:10 -05:00

__init__.py

…

conftest.py

…

test_api_server_process_manager.py

…

test_chat_utils.py

…

test_context.py

[responsesAPI] fix simpleContext streaming output_messages (#34188)

2026-02-09 22:53:07 -08:00

test_grpc_server.py

…

test_responses_utils.py

…

test_ssl_cert_refresher.py

…

test_utils.py

[Bugfix] Treat generation_config max_tokens as default not ceiling (#34063)

2026-02-16 07:58:24 -08:00