Issues

• [enhancement]: Support multiresolution inspection of very large images

  invoke-ai/InvokeAI#9615 · Filter by author kvnloo opened on Oct 1, 2026

• Canvas Shape tool creates a raster layer instead of filling the selected inpaint mask or regional guidance

  bugfrontend
  invoke-ai/InvokeAI#9616 · Filter by author lstein opened on Sep 30, 2026

• Canvas: layer settings (regional prompts, control model) are hidden below the tool settings

  bugfrontend
  invoke-ai/InvokeAI#9617 · Filter by author lstein opened on Sep 30, 2026
  1

• Gallery deletion does not check whether media is still in use

  bugfrontend
  invoke-ai/InvokeAI#9618 · Filter by author lstein opened on Sep 30, 2026

• Possible model-support issues found while documenting model families

  bug
  invoke-ai/InvokeAI#9619 · Filter by author lstein opened on Sep 30, 2026
  2

• [bug]: Z-Image turbo regional guidance.

  bug
  invoke-ai/InvokeAI#9612 · Filter by author Jokerlorian opened on Sep 30, 2026

• FP8 Storage is switched on at install for MXFP8 checkpoints, where it costs quality

  invoke-ai/InvokeAI#9620 · Filter by author Pfannkuchensack opened on Sep 30, 2026

• [enhancement]: Upscaling

  enhancement
  invoke-ai/InvokeAI#9608 · Filter by author valenok-polny opened on Sep 29, 2026
  4

• [bug]: Align Graph.add_edge and validate_self collector type validation

  bug
  invoke-ai/InvokeAI#9610 · Filter by author JPPhoto opened on Sep 26, 2026
  

• [bug]: Nested Iterate execution mixes values across outer iterations

  bug
  invoke-ai/InvokeAI#9609 · Filter by author JPPhoto opened on Sep 26, 2026
  

• [enhancement]: Dynamic Tag Filtering for Images and Videos

  enhancement
  invoke-ai/InvokeAI#9601 · Filter by author luisfilipeac91 opened on Sep 26, 2026

• [bug]: 6.14.1 regression with pytorch_cuda_alloc_conf: backend:cudaMallocAsync — Z-Image bf16 + LoRA takes ~30 min whenever the transformer is (re)loaded (VRAM overflows into Windows shared memory)

  invoke-ai/InvokeAI#9597 · Filter by author cochbild opened on Sep 24, 2026
  2

  

• [enhancement]: Upgrade python from 3.12 to 3.14

  enhancement
  invoke-ai/InvokeAI#9621 · Filter by author aleyan opened on Sep 23, 2026

• [enhancement]: add Qwen-Image-2.1 support

  enhancement
  invoke-ai/InvokeAI#9594 · Filter by author krakotay opened on Sep 21, 2026
  2

• [bug]: Error generating simple image on NVIDIA using FLUX.2 Klein 4B

  bug
  invoke-ai/InvokeAI#9593 · Filter by author gafda opened on Sep 19, 2026
  14

  

• [bug]: Every drag makes the window freeze until refresh since the last update

  bug
  invoke-ai/InvokeAI#9586 · Filter by author Archibaldr opened on Sep 13, 2026
  1

• [bug]: Crashing Krea2 checkpoints after recent update (Anything but the smallest Q8_0 GGUF crashes, Even standard checkpoints)

  bug
  invoke-ai/InvokeAI#9585 · Filter by author CarstenD74 opened on Sep 13, 2026
  9

  

• [bug]: repo_name error when using starter SDQN model

  bug
  invoke-ai/InvokeAI#9580 · Filter by author CubeTheThird opened on Sep 9, 2026
  1

• [enhancement]: Add video duration in seconds with automatic frame count calculation

  invoke-ai/InvokeAI#9576 · Filter by author jimmybrancaccio opened on Sep 7, 2026
  1

• [bug]: Krea2 Inference is slow on MPS (due to hardcoded bfloat16 dtype)

  bug
  invoke-ai/InvokeAI#9575 · Filter by author grunblatt-git opened on Sep 7, 2026

• [enhancement]: Board Groups/Folders

  enhancement
  invoke-ai/InvokeAI#9574 · Filter by author Arvoly opened on Sep 7, 2026

• ai-toolkit DoRA (.magnitude) adapters fail to load for every base except Krea-2

  invoke-ai/InvokeAI#9558 · Filter by author lstein opened on Aug 29, 2026
  

• [bug]: Black images don't work for Inpaint Mask, Images less than 600px can't be used for Flux.2 Ref image

  bug
  invoke-ai/InvokeAI#9553 · Filter by author digitalassassins opened on Aug 27, 2026
  1

  

• [bug]: Segment Anything V2 fails every time

  bug
  invoke-ai/InvokeAI#9551 · Filter by author shanedk opened on Aug 27, 2026
  1

  

• [enhancement]: Default to Flashpack for v7

  enhancement
  invoke-ai/InvokeAI#9548 · Filter by author iwr-redmond opened on Aug 27, 2026
  2