"PHP was created for the browser" funny that it doesn't run in the browservatvall then, in fact id nevervtouches the cloent mschinevat all afaik it's serverecside, did you mean for "thevweb "?

Awesome video and great explanation, thank you!!! ❤🔥 Just in case someone else is having the same problem as I had: "ld hello.o -o hello -lto_library System -syslibroot `xcrun -sdk macosx --show-sdk-path` -e _main -arch arm64" resulted in: "ld: dynamic main executables must link with libSystem.dylib for architecture arm64" I just had to append " -lSystem" at the end and it worked 👍

Thank you so much for this video! It helped me add MCP tools to my node JS local LLM project.

Really interesting and informative! I too tried Image Puzzles on both o1 and Flash Thinking, and both had a tough time: kzbin.info/www/bejne/d2GqhKyneJ6GeJYsi=n1uRaxMq4L6u8Ta5

i wish you do this video step by step.. Because i dont understand anyting

Wow... very good and exploratory analysis. Thank you very much.👍 You reinforce the use specific model to specific problem, for example TOC to Agents... Thanks to remember this!

please integrate it with together api key

Next when you test these models give it an image of an economic calendar from FXStreet and ask it to count the number of Red Bars and Orange Bars in the screenshot or Reading & extracting data from a screenshot of a Rogers and Mayhew Steam Table. I think image analysis is still a critical area these models need to work on.

Which is better at coding, 1206 or flash thinking? I'm not too sure yet about the superiority of CoT for coding in general.

9:20 : why do you believe position 5 is the right answer? O loses: O . . . . X . X . (your starting position) O . . . O X . X . (O places on position 5, what you believe to be the solution) O . . . O X . X X (blocking O, now X has two ways to win, on the third column, and third row) O . O . O X . X X (blocking one of the two ways, that's the best O can do) O . O . O X X X X (X wins) The *actual* best continuation is as follows (there are two possible correct solutions, in position 3, and position 7): O . . . . X . X . (your starting position) O . O . . X . X . (forcing X to block a tris) O X O . . X . X . (only move X can do is to block the tris, threating tris on the second column) O X O . O X . X . (O blocks the tris on the second column, and threatens a tris on both diagonals) O X O . O X . X X (X blocks one of the two) O X O . O X O X X (O wins by doing tris on the other diagonal) This was the continuation for position 3. For position 7 the continuation is similar, in fact it's just rotated. It's all forced moves, so you can quickly verify it yourself. So... Position 9 is wrong, but also position 5 is wrong. The only correct solutions are position 3 and position 7.

good test, we hope to improve the model shortly

what's 254*752-3+(2-7)+5? first convert it to simpler parts then start solve it This model is very good at CoT, so if you make it think step by step, it gives you the right answer. For a better test of this model, you should make it break it down and think about it as much as possible. so the System Prompt should be : always break down any questions into simpler parts, then solve it step by step.

What are your pc specs??

The Ollama modal is quantized to Q4_K_M, so you will lose much quality, especially context. It is not a fair comparison. Also, you may need some guidance to force thinking which o1 and Claude 3.5 are doing but Ollama doesn't support yet.

If you're really good at using the model, could get a lot of work done with it, or does it have too many limitations and weaknesses?

Models from Microsoft are really good

Thanks for this! Also for anyone running into the following error when using ld/linker manually: "ld: warning: ignoring file hello.o, building for macOS-arm64 but attempting to link with file built for macOS-x86_64" 1. remove the hello.o file 2. specify the architecture when running clang ie. clang hello.c -c -o hello.o -arch arm64 3. Then run the command as specified eg. ld hello.o -o hello -lSystem -syslibroot `xcrun -sdk macosx --show-sdk-path` -e _main -arch arm64

FYI, on Mac at least, the BoltAI GUI app will connect to locally-running Ollama and LM Studio models served by those apps

in openweb ui how did you setup a dark mode? it is so useful

please note at 13:46, mini got the answer right, as did llama3.3 at 14:11. i did point out earlier in the video, that position 2 is a correct answer. in the flow of the video i was focused on position 3, even though i had pointed out position 2 was also correct. you've all figured this for yourselves. i'm not editting this video as it doesn't change the narrative. apologies for missing this in the flow of the video

13:46 GPT4o mini got a RIGHT answer. position 2 is a winning move as much as 3 (you said it yourself earlier in the video and you mistakenly discarded the GPT4o mini answer as invalid. Again on 14:11 you wrongly stated Llama 3.3 gets it wrong when it gets it right! You had 1 job! Please fix the video or carry your reputation as a shill.

The answer is always no.

Great video. Liked and subscribed

Hiii, thank your for the excelent tutorial. I was following the tutorial however I saw something strange. The desktop version saws laggy so I added the Stats component to the application. While in the browser I was having 100fps in the application I got 60 fps 😢. The idea of tauri seems very good, but it’s not good for desktop apps since it seems a downgrade of web browser version

What is your hardware configuration ?

Or CLINE with MCP !!

Try maybe lm-studio or vllm also not supporting tools ?

Nemotron got on first try when responding the '254*752-3+(2-7)+5': To evaluate this expression, we need to follow the order of operations, often remembered by the acronym PEMDAS (Parentheses, Exponents, Multiplication and Division from left to right, Addition and Subtraction from left to right). Here's how to break it down: Given expression: `254 * 752 - 3 + (2 - 7) + 5` 1. **Parentheses first**: Solve the expression inside the parentheses. - `(2 - 7) = -5` 2. **Update the expression**: Replace the solved parentheses with its result. - `254 * 752 - 3 - 5 + 5` 3. **Multiplication next** (since there are no exponents): - `254 * 752 = 191008` 4. **Update the expression again**: - `191008 - 3 - 5 + 5` 5. **Finally, handle Addition and Subtraction from left to right**: - `191008 - 3 = 191005` - `191005 - 5 = 191000` - `191000 + 5 = 191005` So, the final result of the expression `254 * 752 - 3 + (2 - 7) + 5` is **191005**.

I think Sam Altman said “no one wants last years model” or the second best model. Unless it’s really faster or better most will stick to one of the others. Good to see though.

No tool calling? Qwen will they be able to?

Google Fi, Phi, but when are they going to release Fo, and Fum? As an Englishman by blood please respond.

What version of OpenWebUI are you using?

ollama run vanilj/Phi-4 Yes?

not yet available on huggingface library searching but if someone has a pull command from ollama please say

I used a Gen10 A2 GPU, and it actually did a really good job on the latest model. It used about 10GB NVRAM at most compared with the o1, and it was not far of a really good model

The 14b parameter model may give worse answers in test time compute as smaller models generally does not reason better vs larger models

does OpenWebUI cost money? can it run *anything* ?

Saved me time and heartache figuring this out the hard way. I can ask for nothing more. Thanks!

Nice video Chris

Bravo!!! 🎉🎉🎉

Are you sure selecting 2 is wrong?

At this point, Supernova Medius is better?🤔

😂 definitely not

you talk too much, can't you hit the point right away

GJ! I want to ask u something, I added the source column but I don't know how to fine tune it..

Excellent stuff, thank you. Would love to see a video on how to use JSONRpc in other projects.

Great job Chris, any plans to support multiple servers like Claude Desktop can? I've built an MCP server that can build Powerpoint decks, I can chain with the SQLite mcp server to pull data sources from a database and dynamically build tables and charts in Powerpoint. All in one prompt

Ollama Need so expensive PC

32b-instruct-q4_K_M - 23gb vram

I was having an error while linking the hello.o file with libraries via ld command, stating System library not found, you can rectify that by using, -lto_library (in place of -l) System (rest all the same) it was really a wonderful video! can't thank you enough Chris!