2 166 613

Evan

evdcush

AI & ML interests

None yet

Recent Activity

liked a dataset 35 minutes ago

anvo25/vlms-are-biased

liked a model 6 days ago

KittenML/kitten-tts-nano-0.1

liked a model 6 days ago

openai/gpt-oss-20b

View all activity

Organizations

liked a dataset 35 minutes ago

anvo25/vlms-are-biased

Viewer • Updated Jul 2 • 11.6k • 476 • 16

liked 2 models 6 days ago

KittenML/kitten-tts-nano-0.1

Updated 8 days ago • 33.7k • 405

openai/gpt-oss-20b

Text Generation • 22B • Updated 4 days ago • 2.37M • • 2.84k

liked a model 12 days ago

black-forest-labs/FLUX.1-Krea-dev

Text-to-Image • Updated 13 days ago • 74.4k • • 634

liked 2 models 19 days ago

mistralai/Voxtral-Mini-3B-2507

5B • Updated 15 days ago • 386k • 519

bosonai/higgs-audio-v2-generation-3B-base

Text-to-Speech • 6B • Updated 15 days ago • 278k • 547

liked a model 22 days ago

moonshotai/Kimi-K2-Instruct

Text Generation • Updated 1 day ago • 481k • • 2.07k

liked a Space 22 days ago

1.01k

UGI Leaderboard

📢

Uncensored General Intelligence Leaderboard

liked a model about 1 month ago

fishaudio/openaudio-s1-mini

Text-to-Speech • Updated Jun 2 • 2.82k • 406

liked 5 datasets about 1 month ago

liked 3 models about 2 months ago

google/magenta-realtime

Updated 5 days ago • 679 • 487

black-forest-labs/FLUX.1-Kontext-dev

Image-to-Image • Updated Jun 27 • 367k • • 2.07k

Menlo/Jan-nano

Text Generation • 4B • Updated Jul 4 • 3.17k • 477

liked a model 2 months ago

mistralai/Magistral-Small-2506

Text Generation • 24B • Updated 15 days ago • 40.5k • 593

reacted to m-ric's post with 🔥 2 months ago

Post

4508

I've made an open version of Google's NotebookLM, and it shows the superiority of the open source tech task! 💪

The app's workflow is simple. Given a source PDF or URL, it extracts the content from it, then tasks Meta's Llama 3.3-70B with writing the podcast script, with a good prompt crafted by @gabrielchua ("two hosts, with lively discussion, fun notes, insightful question etc.")
Then it hands off the text-to-speech conversion to Kokoro-82M, and there you go, you have two hosts discussion any article.

The generation is nearly instant, because:
> Llama 3.3 70B is running at 1,000 tokens/seconds with Cerebras inference
> The audio is generated in streaming mode by the tiny (yet powerful) Kokoro, generating voices faster than real-time.

And the audio generation runs for free on Zero GPUs, hosted by HF on H200s.

Overall, open source solutions rival the quality of closed-source solutions at close to no cost!

Try it here 👉👉 m-ric/open-notebooklm