[FEATURE] SearXNG and multiple ollama inference params

[PieBru]

Hi,
awesome app, KISS simplicity is always a great value!

I would like to suggest a couple of features I found handy in other (not KISS) similar apps.

1. SearXNG, to free from propriatry systems and billing.
2. Multiple Ollama inference params: it would be handy to use differend models, different endpoints, different parameters as the context size.

I use Phi-4 locally, but it would be nice to change to Llama*, Qwen2.5*, or others by simply select from a drop-down list.

Thank you much for this tool!
Cheers from Italy,
Piero

[AnotiaWang]

Hi, thank you for the feedback!

1. SearXNG is just a search service. If we integrate it, we would need to pair it with another service capable of crawling web content (such as Jina, or other self-hosted solutions). I'm afraid that I don't have much time to implement this currently. I recommend checking out this project based on Firecrawl: https://github.com/loorisr/firecrawl-searxng
2. the dropdown simply shows the list of models returned by Ollama. If you have downloaded other models, they should automatically appear in the dropdown.

[AnotiaWang]

Firecrawl has added SearXNG support

[PieBru]

From https://github.com/mendableai/firecrawl/blob/fea249c568c4a8d63bcc2e560fd3655e73350be6/SELF_HOST.md?plain=1#L73

=== /search API ===
By default, the /search API will use Google search.
You can specify a SearXNG server with the JSON format enabled, if you'd like to use that instead of direct Google.
You can also customize the engines and categories parameters, but the defaults should also work just fine.
SEARXNG_ENDPOINT=http://your.searxng.server
SEARXNG_ENGINES=
SEARXNG_CATEGORIES=