👨‍🏫 Tutorial Educational Purpose: FREE gpt-4o-mini, gemma2, claude-3, mixtral, llama3.2 and wikipedia AI

[alist1986]

Para doon sa baguhan, free server po yan ng Open WebUI (formarly Ollama WebUI) na may mga available models na to try and use its features given sa link nya sa github. Ang validity po ng models ay nakadeoende sa rate limits ng models dyan which we don't know. As long as its updated, malamang live yung apis for this server. Sa open-source LLMs lang yung maaaring unlimited api requests kung "models" yung ginamit dyan, or for public unlimited use yung external api na ginamit. Sa commercial, don't expect unlimited for all the public to use. Just maximize as long as it's alive and wait for a reset.
Yung speed of response ng GPT or LLM ay nakadepende sa hardware na ginamit sa server (CPU or GPU), the apis used and speed of your internet connection.

Kung gusto ninyong makagamit ng similar app na ganyan, please visit here and try them.

Spoiler: HF spaces for open webui

Spoiler contents are visible only to Established Members.

Merong may models at merong wala yung iba. Sa wala, it means, you need to add your models and apis individually using the huggingface platform as the open webui is yours to fiddle as you want when you login. Basta may "Workspace" (admin panel) sa kaliwa ng dashboard, pwede ninyong yan gawing personalized UI or just clone it sa HF para separate. Libre naman doon using their CPU hardware. Then use this guide You do not have permission to view the full content of this post. Log in or register now..

Pag wala nyan, fixed UI yan ng gumawa ng server. May mga free to use Open WebUI na ganyan to get you starte d without the hassle of running it. Mabagal ng kaunti since mga online CPU lang yung hardware (na public sa HF), pero works if you know what to avoid - just use apis not models he he. Kayo na bahala sa choices ninyo sa GPT or LLM. But most are the same sa first post ni reonjy sa mga naghahanap din ng libre.

You can still run your own locally using the You do not have permission to view the full content of this post. Log in or register now. link. Nakasaad naman dyan na kailangan ninyo ng Python 11 to avoid compatibility issues. The rest is just following instructions, some background on using similar interfaces and the apis (free or ρáíd) you collected. I'm sure, meron naman dito na hindi lang dependent sa point and click, and don't mind sweating a bit he he. For Educational Purposes naman yung thread, di ba?

Good luck na lang sa inyo.

 

[GingerCup]

Salamat.

 

[alist1986]

Dagdag ko dyan sa post#21 kung sakaling maging separate post ito sa PHC ay bisitahin nyo itong link sa baba. I'm sure, maaring makatulong ito para maging unlimited yung Open WEBUI ninyo or sa 1st post din if its an addition:
You do not have permission to view the full content of this post. Log in or register now.
Yung api sa HF ay libre kung i-deploy ninyo yung inyong chosen models using these links for free:
You do not have permission to view the full content of this post. Log in or register now.
You do not have permission to view the full content of this post. Log in or register now.
You do not have permission to view the full content of this post. Log in or register now.
You do not have permission to view the full content of this post. Log in or register now.

User Tier	Rate Limit
Unregistered Users	1 request per hour
Signed-up Users	50 requests per hour
PRO and Enterprise Users	500 requests per hour

Basahin nyo na lang yung ibang detalye dito sa You do not have permission to view the full content of this post. Log in or register now. Pero sa Groq (1 account), at Openrouter (1 account), sobra-sobra na sa akin using Open WebUI kung isasalang ko ng magdamagang prompts sa research using 1 single LLM model. Kailangan lang na alam mo yung specs/features ng api na gagamitin mo for text, image, video and audio, whether they are multi-modal or just plain text generators. The rest like Google, Cohere, Deepseek, Claude, atbp. na may decent free api requests/day ay halos di na kailangan. Backup na lang for a specific purpose. At least makapag-isip kayo kung alin sa options ng api ang inyong gagamitin kung sakali.
Kung gusto ninyong subukan online, just search "open webui one-click deployment" , "open webui one-click deployment with zeabur" or "deploy open webui in huggingface" para matagal gamitin kung free ang choice ninyo. But a normal laptop/desktop will do using just apis.

 

[NanC-Phc]

Salamat lods

 

[Luffy 19]

Very helpful sa paghahanap ng RRL hahaha

 

[Azt3c_]

Incorrect pass lods

 

[alist1986]

Makuha lang nila yung essence ng post #21 at #24, kahit pa walang Open WebUI, bastante na sila. Nakakahiya man gawin pero kung may account ka sa huggingface, at merong public spaces na open sa gusto mong app like "open webui" na binanggit ko ( na pasok sa free CPU hardware use dyan), sa tabi ng "Running", meron dyan na "Duplicate this space". Click mo lang at intayin matapos. Pagtakbo nyan, lahat ng pinasok nya nasa server mo rin. Dumami na nga ngayon yang "open webui" sa HF spaces. Marami walang laman dahil di pa nila nalagyan ng apis, pero yung iba tulad din ng kay pakner - ready to use. Yung sa pc ko +100 free models kasama yung complete set ng Claude, OpenAI at Google - sari-saring rate limits. Yung mga live remote endpoints ng Ollama (public/private) servers yung extra kung makuha sa web scraping sa search engines o sa mga AI channels/forums.

 

[Cyber Bricked]

ts master

 

[reonjy]

Incorrect pass lods

may palaging nagchichange ng pass, wait.. change back ko, plano ko sana dagdagan ng gpt-4o model kaso dami matigas ulo, nachange pass na to for 3x..

 

[WEIRJD]

support thank you

 

[alist1986]

may palaging nagchichange ng pass, wait.. change back ko, plano ko sana dagdagan ng gpt-4o model kaso dami matigas ulo, nachange pass na to for 3x..

May setting siguro yan sa Admin Panel para di baguhin pakner - di pa yata set yung database mo para multi-login na. Ang alam ko "You do not have permission to view the full content of this post. Log in or register now. to False" sa single user sa solo-mode with no user sa database (no need to login). Pero di siguro yan applied sa case mo dahil enabled yung login at may isang user.
Yan nga nakakainis, may instruction na, kung anu-ano ginagawa he he. Kalimitan dyan, di nagbabasa talaga. May balak pa yatang solohin!
Hinahanda ko na nga thread mo for a graceful exit para yung mga "students" mo na lang mag-deploy ng kanila in case may mangyari tulad nito he he.
Ito malamang yung fix:

DEFAULT_USER_ROLE
Type: str (enum: pending, user, admin)
Options:
pending - New users are pending until their accounts are manually activated by an admin.
user - New users are automatically activated with regular user permissions.
admin - New users are automatically activated with administrator permissions.
Default: pending
Description: Sets the default role assigned to new users.

Lagyan mo na lang ng option na "pending" kapag bago yung user login para madali! Kaya isa lang user ang pwedeng pumasok sa server hangga't di mo payagan yung bago he he.

 

[reonjy]

May setting siguro yan sa Admin Panel para di baguhin pakner - di pa yata set yung database mo para multi-login na. Ang alam ko "You do not have permission to view the full content of this post. Log in or register now. to False" sa single user sa solo-mode with no user sa database (no need to login). Pero di siguro yan applied sa case mo dahil enabled yung login at may isang user.
Yan nga nakakainis, may instruction na, kung anu-ano ginagawa he he. Kalimitan dyan, di nagbabasa talaga. May balak pa yatang solohin!
Hinahanda ko na nga thread mo for a graceful exit para yung mga "students" mo na lang mag-deploy ng kanila in case may mangyari tulad nito he he.
Ito malamang yung fix:

DEFAULT_USER_ROLE
Type: str (enum: pending, user, admin)
Options:
pending - New users are pending until their accounts are manually activated by an admin.
user - New users are automatically activated with regular user permissions.
admin - New users are automatically activated with administrator permissions.
Default: pending
Description: Sets the default role assigned to new users.

Lagyan mo na lang ng option na "pending" kapag bago yung user login para madali! Kaya isa lang user ang pwedeng pumasok sa server hangga't di mo payagan yung bago he he.

ok na yung Database, enable ko after dinner yung multi-user.. tas dagdagan ko isang model na naka fine-tuned for educational..

Edit:
UPDATE: enabled user signup

 

[alist1986]

ok na yung Database, enable ko after dinner yung multi-user.. tas dagdagan ko isang model na naka fine-tuned for educational..

He he. Medyo kumplikado rin yung setup ng Open Webui, pero lahat naman pwedeng gawin sa UI kaysa manu-manuhin mo sa config/env. Kinabisado ko rin yung iba like LMStudio, Jan, Lllama.cpp, vLLM, gpt4all, atbp. para ma-training utak ko sa kung alin yung ok. Mas madaling mag-deploy ng actual llm models gamit ang HF. kusang bumubukas. Sa tingin ko, apis yung gamit mo dyan sa server mo ng binuksan ko noong isang araw.
Pero yan ang the best UI in terms of features. Para kang may Coze.com sa loob ng pc mo using the Tools integration (function calls) na specific para sa isang model using your Web Search, image generation, at yung stated na web scraping, atbp. Para sa premium users kung sakali he he. Bilib ako. Pero mataas kumain ng memory pag umandar sa'kin.
Yung defaults naman sa open webui, pag enabled mo dyan sa server ay sapat na for normal users using your decent LLM models. Tingnan ko kung kaya kong makita yung api ng server mo (with permission) para magamit externally. May feature yan kasi na wala sa docs na nabasa ko sa kanilang discussions. Kailangan lang ng matinong log viewer. Yan kasi yung trick sa ibang UIs and websites. Nakukuha sa trial and error he he.

 

[Zitzuki]

pwedeng pasabit po thank you

 

[reonjy]

He he. Medyo kumplikado rin yung setup ng Open Webui, pero lahat naman pwedeng gawin sa UI kaysa manu-manuhin mo sa config/env. Kinabisado ko rin yung iba like LMStudio, Jan, Lllama.cpp, vLLM, gpt4all, atbp. para ma-training utak ko sa kung alin yung ok. Mas madaling mag-deploy ng actual llm models gamit ang HF. kusang bumubukas. Sa tingin ko, apis yung gamit mo dyan sa server mo ng binuksan ko noong isang araw.
Pero yan ang the best UI in terms of features. Para kang may Coze.com sa loob ng pc mo using the Tools integration (function calls) na specific para sa isang model using your Web Search, image generation, at yung stated na web scraping, atbp. Para sa ρrémíùm users kung sakali he he. Bilib ako. Pero mataas kumain ng memory pag umandar sa'kin.
Yung defaults naman sa open webui, pag enabled mo dyan sa server ay sapat na for normal users using your decent LLM models. Tingnan ko kung kaya kong makita yung api ng server mo (with permission) para magamit externally. May feature yan kasi na wala sa docs na nabasa ko sa kanilang discussions. Kailangan lang ng matinong log viewer. Yan kasi yung trick sa ibang UIs and websites. Nakukuha sa trial and error he he.

uu, own endpoint gamit ko diyan pakner.. pinasok ko sa one-api para isang api lang sa lahat ng llms..

You do not have permission to view the full content of this post. Log in or register now.

 

[WEIRJD]

possible po ba magkaroon ng claude sonnet 3.5?

 

[reonjy]

possible po ba magkaroon ng claude sonnet 3.5?

yes, trabahuin ko mamaya or bukas..

 

[F-Society]

How to register po?

 

[reonjy]

How to register po?

ayy sori, try ulet.. ok na.. enabled na..

 

[F-Society]

ayy sori, try ulet.. ok na.. enabled na..

kala ko may hidden hidden ee haha thank you sir