I have added LiteLLM to the App Wishlist, please upvote!
https://forum.cloudron.io/topic/13141/litellm-openrouter-self-hosted-alternative-proxy-provides-access-to-openai-bedrock-anthropic-gemini-etc?_=1737171723819

Paging @cpa

Hi @Lanhild sorry to hijack this thread, but can you please have a look at packaging LiteLLM please?
LiteLLM already has Langfuse built in, so this might be interesting to you.

Link to App Wishlist: https://forum.cloudron.io/topic/13141/litellm-openrouter-self-hosted-alternative-proxy-provides-access-to-openai-bedrock-anthropic-gemini-etc?_=1737171723819

LiteLLM was earlier integrated into OpenWebUI and was marvellous.
It was removed recently and I had posted about it here: https://forum.cloudron.io/topic/11957/litellm-removed-from-openwebui-requires-own-separate-container

Now, having successfully tested the docker, wish it could be on Cloudron and can act as proxy for all of our AI applications that require API access to any AI provider.

This would be fantastic - one application to rule them all!

---

• LiteLLM: LLM Gateway on Cloudron - provide model access, logging and usage tracking across 100+ LLMs. All in the OpenAI format.

---

• Main Page: https://www.litellm.ai/
• Git: https://github.com/BerriAI/litellm/
• Licence: AGPL
• Docker: Yes, Docker Packages
• Demo: 7 Day Enterprise Trial
• Discussion/Community: Discord

---

• Summary: LiteLLM exposes an OpenAI compatible API that proxies requests to other LLM API services. This provides a standardized API to interact with both open-source and commercial LLMs.

This can be a self-hosted alternative to OpenRouter.
Any application - including OpenWebUI (on CLoudron) can use this OpenAI compatible API endpoint to access over 100 integrations from OpenAI, Amazon Bedrock, Anthropic models, Google Gemini models, Grok, Deepseek, etc.

• 100+ LLM Provider Integrations
• Langfuse, Langsmith, OTEL Logging
• Virtual Keys, Budgets, Teams
• Load Balancing, RPM/TPM limits
• LLM Guardrails

---

• Notes: This would make any of the AI on cloudron or outside (e.g. Khoj) very powerful, bridge the divide between models as they are released and afford us access easily through a single API.

This is superpower, AI on steroids, giving us access to leverage the comparive advantage of each AI model (e.g. Claude for programming, Deepseek for cheapness and Bedrock for longer context window and memory, etc.

---

• Alternative to / Libhunt link: OpenRouter.ai
• Installation Instructions: https://docs.litellm.ai/docs/proxy/docker_quick_start

I have this backup issue with two instances.
In one with Nextcloud, it is empty directories.
In a Wordpress one, it is executable files.
Hoping to tackle this two this weekend.

Don't understand why the devs will devote time to these side quests that won't be used much at all, when everybody is already in a slack / discord or other forum.

Hi, Can we have some idea of how many times, various apps are installed in the cloudronverse? Can be an approximate number too - below 100, 100 to 500, 500 to 10000, 10K to 50K, etc.

If we can track this over time, we can find changes in popularity and other trends too.

Also, we would love to see a ranking by number of installations to find the most popular and least popular apps.

Thank you!

@girish Beg to differ, this is not just another notes/link/bookmark app!!

This will be so useful for those who collect a lot of stuff from around the internet - researchers and such.
Once you use something like Mymind which automatically organises your messy collections, you will never go back.

Reminds me of one more thing I wanted to ask as a feature request, will post one there too: Can we have statistics of how many installations of each of the apps are in the cloudron ecosystem (surely you must have those stats), and can we rank them by popularity please, somewhere?

Thank you!

@girish I did run that, and got a very long list of files - php, png, svg on a wordpress site along with all the plugins.

Absolutely no idea how to proceed.

But for now, I will disable the apps shown the list and try, thank you!

This is an alternative to raindrop.io too.

@girish sorry for the dumb question, but where do I run the 'find' command? I tried running it in the terminal of the nextcloud/wordpress app, but I got the 'no such file or directory' error.

Mine is running since 2019 and I am a big fan!

I am getting this error too, as well as this new error in another of my cloudron:
QUote:
Backup failed: Too many executable files. Run "find /home/yellowtent/appsdata/8679c3b3-59b6-4d7e-a84a-d11dab4d5a19 -type f -executable" to investigate.

---

How can I solve both the empty directories as well as this error please? Thank you!

---

• Paperless-AI: To automatically analyse and tag documents in Paperless-NGx

---

• Main Page: https://github.com/clusterzx/paperless-ai/?tab=readme-ov-file
• Git: https://github.com/clusterzx/paperless-ai/
• Licence: AGPL
• Docker: Yes

---

• Summary: Leverage AI through OpenAI or Ollama Locally to analyse all your documents and tag them - this will help in total automation!

• Features:

Automatic document scanning in Paperless-ngx

AI-powered document analysis using OpenAI API and Ollama (Mistral, llama, phi 3, gemma 2)

️ Automatic title, tag and correspondent assignment

---

• Notes: Its about time we had a handsoff approach to document handling. Tagging documents is such a PITA and any automation with intelligence would be most welcome!

---

• Installation Guide: https://github.com/clusterzx/paperless-ai/wiki/Installation
• Screenshots:
  
  

Great alternative to Clerk, Auth0 and like.
More features than Firebase too.

If we can get this on cloudron it would be great - we can use it as an authentication backend for all other apps.

Nova is here!

Cloudroners, unless you are living under the rock, you might have been hit with the launch of the Amazon Nova - that is matching or beating existing frontier models and is multimodal from the start go.

So, naturally, the only next question is, how do we get access to this emergent model at the lowest cost and ease of use.

OpenWebUi, fortunately has been a great boon. I have typingmind and half a dozen other AI Chat Wrapper Apps, and none of them are as powerful as OpenWebUi - but am preaching to the choir here.

OpenWebUi is, as per my searches, currently, the easiest way to directly access the cheap Nova. What we need is the Amazon Bedrock Access Gateway that we can add to the OpenWebUi that also allows us to access not only Nova, but also Anthropic, Meta, Cohere, Stability and others.

What is Bedrock?

From online:
Amazon Bedrock is a fully managed service that provides access to a wide range of high-performing foundation models (FMs) from leading AI providers like Anthropic, Cohere, and Stability AI. This allows you to select the best model for your specific use case, whether you’re analyzing internal documents, questioning financial records, or processing customer information. Importantly, Amazon Bedrock is designed with a focus on security, privacy, and responsible AI practices, ensuring your sensitive data remains protected.

Here are some reading on how to connect OpenWebUi with Bedrock and other details:

https://gauravve.medium.com/connecting-open-webui-to-aws-bedrock-a1f0082c8cb2
https://jrpospos.blog/posts/2024/08/using-amazon-bedrock-with-openwebui-when-working-with-sensitive-data/

---

Please upvote my entry to the App Wishlist to get Bedrock on Cloudron.
https://forum.cloudron.io/topic/13032/amazon-aws-bedrock-access-gateway-for-connecting-with-webopen-ui-other-applications

Thank you!

---

• Title: Amazon AWS Bedrock Access Gateway - for connecting with WebOpen UI

---

• Main Page: https://aws.amazon.com/bedrock/
• Git: https://github.com/aws-samples/bedrock-access-gateway/blob/main/src/Dockerfile_ecs
• Licence: ? AGPL ? (I am not sure)
• Docker: Yes
• Demo: https://aistylist.awsplayer.com/

---

• Summary: Now that Amazon Nova is performing at par with Claude Sonnet 3.5 and beating all other leading models from OpenAI, Google Gemini, we need the AWS Bedrock Access Gateway so that we can connect our OpenWebUI to AWS and take advantage of the multimodal capabilities of Nova at low prices (apparently 75% cheaper)

---

• Notes: The AWS Bedrock Access Gateway also provides access to all models of Anthropic, Cohere, as well as Meta's (Llama2/3), Mistral/Mixtral, etc at cheaper cost.
• Running Ollama locally is not ideal since most of us don't run our Cloudron on servers with GPU. Instead, we can access Ilama at great cost savings as well as privacy by using AWS Bedrock's Llama.
• The application is very stable and enterprise ready (Amazon quality assurance) - so we can jump in right away without waiting for bug fixes and such.

---

• Tutorial on how to connect Bedrock Access Gateway to OpenWebUI: https://jrpospos.blog/posts/2024/08/using-amazon-bedrock-with-openwebui-when-working-with-sensitive-data/

---

• Screenshots:

Can we please consider this?
The virtual servers we run are hardly suited for good Ollama output anyways. Now that the tokens cost for lower models of OpenAI, Grok, Anthropic, Google, etc are low, there is actually no point to have Ollama bundled.

Pipelines however, add a great deal of flexibility and functionality to OpenWebUI and can enhance it to no limit!

Thank you

Upvote for whitelabelling the branding.

Can't wait to get this on the store. We need a nice rounded set of AI Apps on Cloudron, now that AI is exploding with hundreds of wrappers and apps...

As well as OpenAI's next 12 days!