NFTCID

AI & ML interests

None yet

Recent Activity

reacted to csabakecskemeti's post with 🔥 about 13 hours ago

Just sharing a result of a homelab infrastructure experiment: I've managed to setup a distributed inference infra at home using a DGX Spark (128GB unified gddr6) and a linux workstation with an RTX 6000 Pro (96GB gddr7) connected via 100Gbps RoCEv2. The model I've used (https://lnkd.in/gx6J7YuB) is about 140GB so could not fit either of the GPU. Full setup and tutorial soon on devquasar.com Screen recording: https://lnkd.in/gKM9H5GJ

liked a model about 13 hours ago

Qwen/Qwen3-TTS-12Hz-1.7B-CustomVoice

liked a dataset about 13 hours ago

facebook/action100m-preview

View all activity

Organizations

None yet

reacted to csabakecskemeti's post with 🔥 about 13 hours ago

Post

3205

Just sharing a result of a homelab infrastructure experiment:

I've managed to setup a distributed inference infra at home using a DGX Spark (128GB unified gddr6) and a linux workstation with an RTX 6000 Pro (96GB gddr7) connected via 100Gbps RoCEv2. The model I've used (https://lnkd.in/gx6J7YuB) is about 140GB so could not fit either of the GPU. Full setup and tutorial soon on devquasar.com

Screen recording:
https://lnkd.in/gKM9H5GJ

3 replies

liked a model about 13 hours ago

Qwen/Qwen3-TTS-12Hz-1.7B-CustomVoice

2B • Updated 3 days ago • 91.8k • 536

liked a dataset about 13 hours ago

facebook/action100m-preview

Viewer • Updated 12 days ago • 120k • 3.31k • 121

liked a Space 1 day ago

Qwen Image Multiple Angles 3D Camera

🎥

1.16k

Adjust camera angles in images using 3D controls or sliders

liked 2 models 1 day ago

microsoft/VibeVoice-ASR

Automatic Speech Recognition • 9B • Updated 5 days ago • 46.9k • 554

nvidia/personaplex-7b-v1

Audio-to-Audio • Updated 4 days ago • 34.9k • 1.06k

liked 2 models 10 months ago

manycore-research/SpatialLM-Llama-1B

Text Generation • 1B • Updated Mar 21, 2025 • 166 • 992

nvidia/GR00T-N1-2B

Robotics • 2B • Updated Sep 2, 2025 • 232 • 344

liked 2 models 11 months ago

sesame/csm-1b

Text-to-Speech • Updated Dec 1, 2025 • 43.5k • 2.32k

CohereLabs/c4ai-command-a-03-2025

Text Generation • 111B • Updated Oct 30, 2025 • 3.58k • • 377

reacted to fdaudens's post with 👍 11 months ago

Post

3526

Trying something new to keep you ahead of the curve: The 5 AI stories of the week - a weekly curation of the most important AI news you need to know. Do you like it?

For more AI stories and deeper analysis, check out my newsletter: https://open.substack.com/pub/fdaudens/p/ai-competition-heats-up-grok-3-iphone

1 reply

liked a model 11 months ago

deepseek-ai/DeepSeek-R1-Zero

Text Generation • 685B • Updated Mar 27, 2025 • 3.02k • 941

upvoted a paper 12 months ago

Goku: Flow Based Video Generative Foundation Models

Paper • 2502.04896 • Published Feb 7, 2025 • 106

liked a dataset 12 months ago

saiyan-world/Goku-MovieGenBench

Viewer • Updated Feb 11, 2025 • 1k • 679 • 215

reacted to Xenova's post with 👍 12 months ago

Post

14584

We did it. Kokoro TTS (v1.0) can now run 100% locally in your browser w/ WebGPU acceleration. Real-time text-to-speech without a server. ⚡️

Generate 10 seconds of speech in ~1 second for $0.

What will you build? 🔥
webml-community/kokoro-webgpu

The most difficult part was getting the model running in the first place, but the next steps are simple:
✂️ Implement sentence splitting, allowing for streamed responses
🌍 Multilingual support (only phonemization left)

Who wants to help?

13 replies

liked a model 12 months ago

deepseek-ai/DeepSeek-R1

Text Generation • 685B • Updated Mar 27, 2025 • 389k • • 13k

reacted to m-ric's post with 👀 about 1 year ago

Post

1462

𝗠𝗶𝗻𝗶𝗠𝗮𝘅'𝘀 𝗻𝗲𝘄 𝗠𝗼𝗘 𝗟𝗟𝗠 𝗿𝗲𝗮𝗰𝗵𝗲𝘀 𝗖𝗹𝗮𝘂𝗱𝗲-𝗦𝗼𝗻𝗻𝗲𝘁 𝗹𝗲𝘃𝗲𝗹 𝘄𝗶𝘁𝗵 𝟰𝗠 𝘁𝗼𝗸𝗲𝗻𝘀 𝗰𝗼𝗻𝘁𝗲𝘅𝘁 𝗹𝗲𝗻𝗴𝘁𝗵 💥

This work from Chinese startup @MiniMax-AI introduces a novel architecture that achieves state-of-the-art performance while handling context windows up to 4 million tokens - roughly 20x longer than current models. The key was combining lightning attention, mixture of experts (MoE), and a careful hybrid approach.

𝗞𝗲𝘆 𝗶𝗻𝘀𝗶𝗴𝗵𝘁𝘀:

🏗️ MoE with novel hybrid attention:
‣ Mixture of Experts with 456B total parameters (45.9B activated per token)
‣ Combines Lightning attention (linear complexity) for most layers and traditional softmax attention every 8 layers

🏆 Outperforms leading models across benchmarks while offering vastly longer context:
‣ Competitive with GPT-4/Claude-3.5-Sonnet on most tasks
‣ Can efficiently handle 4M token contexts (vs 256K for most other LLMs)

🔬 Technical innovations enable efficient scaling:
‣ Novel expert parallel and tensor parallel strategies cut communication overhead in half
‣ Improved linear attention sequence parallelism, multi-level padding and other optimizations achieve 75% GPU utilization (that's really high, generally utilization is around 50%)

🎯 Thorough training strategy:
‣ Careful data curation and quality control by using a smaller preliminary version of their LLM as a judge!

Overall, not only is the model impressive, but the technical paper is also really interesting! 📝
It has lots of insights including a great comparison showing how a 2B MoE (24B total) far outperforms a 7B model for the same amount of FLOPs.

Read it in full here 👉 MiniMax-01: Scaling Foundation Models with Lightning Attention (2501.08313)
Model here, allows commercial use <100M monthly users 👉 MiniMaxAI/MiniMax-Text-01