Automated AI Web Researcher Ollama - Install Locally for Free Research

[@cynicalcamel6143]

Haha I made this program over a couple of months and uploaded it like 3 days ago, and someone already made a video on it! Thanks for helping out everybody mate much appreciated!

[@fahdmirza]

Awesome job on creating it!

[@smileshs1]

In the morning, I woke up to a reddit notification about the same thing and when I saw it, I was like, I don't know coding so let's leave it. But here you are, you've already made a video tutorial on this, which seems so apt for me. Thanks & Good luck for your channel man. Godspeed!

[@smileshs1]

Is it possible to do a literature review with this if I have a bunch of research papers ? Or any other solution I mean. It'd be great.

[@bomsbravo]

phi 3 14b you created Pi! lol Amazing video. Your videos are getting better each day. Thank you for your hard work and generosity.

[@makerspersona5456]

Great video! You’re making a big difference.

[@fahdmirza]

Thanks, I appreciate it.

[@AKBARESFAHANI]

Thank you for showing clearly how to expand the context size. Ollama only does 2048 from the box

[@scottx1415]

I appreciate you putting this out there, but wait: It took a while to setup It ran a long time It gave a very short response 99% sure I would have gotten a better response based on my extensive testing with the latest Chatgpt models, which are a reasonable flat rate (or even the older free ones) Did I miss something? Is this just about saving a bit of $s - good, but how about a quality of the output comparison?

[@cynicalcamel6143]

Did you increase the context in the modelfile like indicated in the github instructions? For some reason Ollama reverts no matter what to like 2k context so your garuanteed short responses unless you make a custom model as shown in the github instructions

[@TheShreyas10]

Hi Fahd interesting video and thanks for the update, I saw this app yesterday on Reddit and its really amazing work. I have one request can you make a video on the best tool or repo out there specifically towards extracting tabular info from PDF, especially the unstructured PDFs. Thanks in advance.

[@johncarter619]

Miner U or docling...

[@ElChapoDel8]

Would be amazing to integrate it with openweb ui

[@fahdmirza]

Indeed

[@lalpremi]

thank you, have a great day.

[@fahdmirza]

You too, thanks for watching!

[@dracarys9]

System initialization failed: LLM test failed: Ollama API request failed with status 404: {"error":"model 'custom-phi3-32k-Q4_K_M' not found"}

[@pendarhadinezhad6084]

great, but in the last step i got this error: No module named 'termios'

[@pleabargain]

Great demo but install fails (after 45minutes of trying) on win11. Maybe the devs will figure out a less resource intensive build process for Win11.