SuperPat

The only workaround is to use the GGML models version of Stable Diffusion 3.5 or Flux.1-dev from the gallery Unfortunately, the GGML models version of Flux.1-schnell (for its compatible commercial...

Mistral API platform return a capabilities object in their /models API return object: ``` "capabilities": { "completion_chat": true, "completion_fim": false, "function_calling": true, "fine_tuning": false, "vision": false, "classification": false }, ```...

With abliteratedv2, throw error: > Unexpected err=OutOfMemoryError('CUDA out of memory. Tried to allocate 90.00 MiB. GPU 0 has a total capacity of 31.36 GiB of which 19.62 MiB is free....

I found a quantized GUFF version of this model: https://huggingface.co/city96/FLUX.1-schnell-gguf/tree/main Are they compatible with LocalAI and if yes how to use it?

The model gallery contains a FLUX.1 DEV ggml models that works, but not the schnell version. We should add this one: https://huggingface.co/city96/FLUX.1-schnell-gguf/resolve/main/flux1-schnell-Q2_K.gguf

Mistral-small-3.1 with vision is now supported in ollama in this PR: https://github.com/ollama/ollama/pull/10099

Same with RTX 5090, This error appear because RTX5000 need cu128 Code need to be upgraded to Pytorch 2.7.0 with CUDA12.8

_ Using models not available in /chat/completions _ The deep research tool _ The code interpreter tool (for Excel file analysis and its ability to create file with python code)

Models available only on /responses API: https://platform.openai.com/docs/models/gpt-5-pro https://platform.openai.com/docs/models/gpt-5.1-codex https://platform.openai.com/docs/models/o1-pro ....

I am very interested in the support of vision models in localAI particularly [Llama-3.2-11B-Vision](https://huggingface.co/meta-llama/Llama-3.2-11B-Vision) and [Pixtral-12b](https://huggingface.co/mistral-community/pixtral-12b-240910)