Видео

It takes a long time, an hour and a third, and it is not fixed.

It's Flux Schnell fine-tuned, that's why it has Flux-like quality. 😁 Have you found a Github to install it local without comfy? I don't like comfy.

on my way to test the GGUF , on my rtx 3070 , and will provide you my feedback , thank you for this quick news you rock !

I think I lost the directions. 🤔 I have two questions around the video subject: If we set this value in modelfile to 6144 (6K) tokens using a model like Phi-3-mini-4k-instruct which support only 4096 (4k) tokens, the resulted model will support a window context size of only 4096 (4k). Is that true? Also, do we need to create a new model using modelfile with new num_ctx parameter value, if we are using the ollama api which support sending options like: "options": {"temperature": 0, "num_ctx": 5120} and in this case, does the model obey the modelfile parameter value or the value sent with the request? Thanks for the good content🌹🌹🌹

Great Video. Please add the links to the description bro

I've tried declaring or selecting the "claude-3.5-sonnet" model and ask it "what company were you created by?" and it responds with "OpenAI" - lol! So safe to say, this isn't really accessing the Sonnet model. Anyone know what's going on or how to fix?

I downloaded this on my phone! And it is really nice! Like personable, I wonder how they did it. It is exviting.

i paste my token into the box but it did'nt show any thing:))

Can you please show how to easily expose the ollama api when running in colab this way?

Thanks

What’s your opinion about this for managing memory within a session?

I think people are misunderstanding this app. It allows you to use natural language to execute ffmpeg commands. Which is really good and really powerful. But you need to know what ffmpeg can do to set expectations and get a good outcome. I am hopeful more apps like this will come out that lower the barrier for using tech. Here are 2 cool examples people can try: 1)Upload an image and some audio. Have it put the waveform of the audio over the image in a new video. 2)Take a static image and convert it to an mp4 file. Useful if you are doing lipsync. Bonus one: Take an mp4 and extract the audio as an mp3.

Seeling 10k credits codes on cheap

is there an api where we could send a image generated by for example flux to create a slideshow? I want to try it but I dont see any synergy.

well done this collab stuff is so good for so many, im proud to see you do these for people

@Fahd & @Eng.Mohammad, I have increased my Text/input, Tokens & Context input to 5Milloin !!! I have Huge files in 100mb - 25GB !! .PDF, Images, RAGS, agents !! I am now writing HUGE FULL programs. Claude, Omni-01, gpt4o & ALOT MORE ALL UNLOCKED FOR FREE ... UNLIMITED TEXT-TOKENS ....Wooohoooo 😄😄😄😄👍😎

good to know 🥰 can you do anything else in custom file?

do you have any idea how can I share this agent with collagues intern in our office ?

extremely useful, thanks!

hi bro fahd ,, does incrsing the context lenght make me uplolad bigger PDF for RAG ,,,thanks for all your work .

Thank you for these awesome videos.

Thank You Fahd 💪

Is there a character limit for test-to-speech generation?

thank you. Do you know how to save the output .webp as a video file so you can import into a video editor? I tried saving the .webp but it is just a static image. Was thinking there is a comfy UI that might be able to do this

Yes, but to imagine that distant future i recommend new french animation movie Mars Express (2023), the rare sci-fi which include all recent trends, including hidden ones like brain "farming" (human brains torrents to hire as cheapest processing unit on planet which uses only sugar), jailbroken robots and etc.

The analogy of AI jobs and traditional jobs is like that of Automobile and Bullock cart. I think people will have to find that job satisfaction by learning AI. People cannot have satisfaction of riding a Bullock cart in 20th century.

Greats picks, would be helpful, if you add links to these spaces in your video description. Thx

give me the code that shall copy into the colab.research.google?

Cool video thanks, any easy way to connect externally? gradio interface?

You shouldn't have shared this. No one can thank you enough! ;) A search will bring up a VScode AI tool by the same name. Looks cool as well, but I do not think they are related. Thanks.

i m so excited for this just what we open source community needed to move ahead do show us how to improve the prompts with llm so that it generates error free

Qwen2.5 Coder 32B is also finetuned for other tasks like translation. i do not know why but Coder was better then Qwen2.5 32b. but absloulently amazing. generaly whatever you suggest is very good level. Because you isntalling and trying yourself. thank you for your efforts.

If it has the capability to handle both NoSQL and vector databases, as you mentioned, I might consider migrating to 23AI. Cool video as always, thanks!

Prompt outputs failed validation CLIPLoader: - Value not in list: clip_name: 't5\google_t5-v1_1-xxl_encoderonly-fp8_e4m3fn.safetensors' not in ['clip_l.safetensors', 't5xxl_fp16.safetensors', 't5xxl_fp8_e4m3fn.safetensors', 't5xxl_fp8_e4m3fn_scaled.safetensors'] DownloadAndLoadMochiModel: - Value not in list: vae: 'mochi_preview_vae_bf16.safetensors' not in ['mochi_preview_vae_decoder_bf16.safetensors'] ???

great video! i was an oracle DBA for 30 years starting with version 5.1.

Thank you for the tutorial! :) How long can be the video made from image? Is it possible to change the length?

I am a subscriber on your channel since last 2 months and I have learnt a lot. Thank you for all the great content. 👍

Please make a video about how to use the cloud infrastructure for running AI workloads like comfyui, ollama etc. Please also share guidance, tips and tricks to get things done at the cheapest cost (from Indian perspective).

How do u change the webcam to external one like Logitech

nice! how can i install and use this to mask my videos?

Thank you very much, the gui deployment has always been problematic, this video has resolved the issue that has been troubling me for a long time

Does suno still work? They’ve upped their security recently..

hi, is it still free

I got a GTX 4070 Super with 12GB VRAM, Ryzen 7950x and 32G Ram. Can I run these models with this setup? (I am still in basic learning phase, so I am not yet ready to use cloud servers from AWS or MAssed Compute etc.)

Thank you for your kind guidance always. I am using a MacOS(MPS). I installed the nodes and models using the procedure you taught me, but the following error occurred. I tried again many times, but it did not work. Please let me know the cause and solution of this. Warning: Missing Node Types When loading the graph, the following node types were not found: LTXVConditioning LTXVScheduler LTXVlmgToVideo

so the first thing is to use Ubuntu Desktop ( Linux ) instead of Windows. No Windows or Mac OS?

it doesnt work on 6800 xt

I like this better than local install. Easier to start up. Question how do you do updates?

Thanks

Does runway offers open source models for this usecase ?