Batch encode wont move on to next image after last update

Nobody has claimed this yet.

After this morning's update (thanks for the lora support in nf4!) Batch encoding wont move on after image 1.
It just sits there being stuck:

[Memory Management] Current Free GPU Memory: 9592.02 MB
[Memory Management] Required Model Memory: 319.75 MB
[Memory Management] Required Inference Memory: 1024.00 MB
[Memory Management] Estimated Remaining GPU Memory: 8248.27 MB
Moving model(s) has taken 3.06 seconds
Skipping unconditional conditioning when CFG = 1. Negative Prompts are ignored.
To load target model JointTextEncoder
Begin to load 1 model
[Memory Management] Current Free GPU Memory: 9264.32 MB
[Memory Management] Required Model Memory: 5227.11 MB
[Memory Management] Required Inference Memory: 1024.00 MB
[Memory Management] Estimated Remaining GPU Memory: 3013.21 MB
Moving model(s) has taken 1.90 seconds
Distilled CFG Scale: 3.5
token_merging_ratio = 0.2
To load target model KModel
Begin to load 1 model
[Memory Management] Current Free GPU Memory: 4660.20 MB
[Memory Management] Required Model Memory: 6246.84 MB
[Memory Management] Required Inference Memory: 1024.00 MB
[Memory Management] Estimated Remaining GPU Memory: -2610.64 MB
[Memory Management] Loaded to Shared Swap: 3448.26 MB (asynchronous method)
[Memory Management] Loaded to GPU: 2798.50 MB
Moving model(s) has taken 60.94 seconds
0%| | 0/30 [00:00

No source file or test is named. Start by reproducing batch encoding after the update with LoRA support in nf4 and the reported Queue, Async CPU, and Shared options, then inspect the logged model-loading and inference transition. Done means batch encoding advances beyond the first image and can be interrupted or skipped without a hard reboot.

Written by the indexing model from the issue text.