kingbri 43f9483bc4 Model: Add tensor_parallel_backend option ...

This allows for users to use nccl or native depending on the GPU setup.
NCCL is only available with Linux built wheels.

Signed-off-by: kingbri <8082010+kingbri1@users.noreply.github.com>

2025-08-17 22:35:10 -04:00

..

core

Model: Add tensor_parallel_backend option

2025-08-17 22:35:10 -04:00

Kobold

Tree: Format

2025-05-17 00:46:40 -04:00

OAI

API: Persist request IDs and append full_text to finish chunk

2025-07-25 12:27:44 -04:00

server.py

Args: Expose api-servers to subcommands

2025-02-10 23:39:46 -05:00