Chris Bruner comments

Results 96 comments of


                                            Chris Bruner

cuda device unavailable error results in failed memory update leading to concurrent model load when no space actually available

``` chris@FORGE:~$ ollama ps NAME ID SIZE PROCESSOR UNTIL llama3.2:1b baf6a787fdff 2.7 GB 100% GPU 4 minutes from now chris@FORGE:~$ ollama run llama3.2:latest Error: llama runner process has terminated: CUDA...

Segfault on opening

Yes that's the same issue.

Cmake error

removing the line that contains qt6_add_ui get's by that problem as widgets are imported by the # Qt6 Setup find_package(Qt6 REQUIRED COMPONENTS Widgets)

Cmake error

Also SavingThread.cpp has #include "SavingThread.h " Note the space after .h #include "ui_CameraConnectDialog.h" in CameraConnectDialog.h is wrong. Was not able to get to compile even with these fixes.

Cmake error

Sorry, it didn't work. ``` git reset --hard HEAD HEAD is now at f477c6a Update README.md chris@FORGE:~/github/Live-Video-Magnification$ git pull Updating f477c6a..d83a55b Fast-forward CMakeLists.txt | 10 +++++----- CMakePresets.json | 71 +++++++++++++++++++++++++++++++++++++++++++++++++++++++++--------------...

Cmake error

[CMakePresets.json](https://github.com/user-attachments/files/18415843/CMakePresets.json)

Cmake error

PR done!

Now get's list of models from ollama instead of hardcoding them. also Improve Browser navigation and error handling

Mainly for making Ollama more reliable but probably others as well.

Now get's list of models from ollama instead of hardcoding them. also Improve Browser navigation and error handling

**This is much more stable** At this point using ollama I am able to run the default agent task (opening openai and clicking the first link). It's also fairly fast....

Now get's list of models from ollama instead of hardcoding them. also Improve Browser navigation and error handling

You should be able to do three separate pull requests. (I don't know why they were all concatenated).