Boyang ZHONG dc72a35f4e [bugfix] Return individual images for batched image API requests (#1917) ...

For n > 1, the image endpoints returned one preview grid instead of n images.
The endpoints now run one model batch, save each decoded sample as its own file,
and return one URL or base64 item for each image. This applies to generations
and edits.

A keyword-only return_samples field (default False) is added to OutputConfig,
SamplingParam and ForwardBatch. Batched image requests set it and do not build
preview frames. The worker keeps the sample tensor for these requests, and
non-output ranks clear the field. Defaults and positional arguments do not
change.

2026-10-07 15:06:02 -07:00

..

installation

[kernel] Build + allow attn_qat_infer FP4 attention on sm_121a (DGX Spark) (#1598)

2026-10-07 04:09:39 -07:00

installation.md

[docs]: FastH3 V2 cookbook serving and MLX install guide (#1884)

2026-09-29 13:03:18 -07:00

quick_start.md

[docs]: add a maintained inference cookbook (#1290)

2026-08-21 01:53:44 -05:00

v1_api.md

[bugfix] Return individual images for batched image API requests (#1917)

2026-10-07 15:06:02 -07:00