Loading video...

Video Failed to Load

There was a problem loading this video. This could be due to a temporary network issue or the video might be unavailable.

Our first open-source release. YOLO26-MLX, native YOLO26 on Apple Silicon. No PyTorch. No external GPU. Up to 2.6x faster inference. Up to 1.7x faster training. Accuracy within 0.2% of official results. It won't be the last.

webAI

5,655 subscribers

97,203 views • 4 months ago •via X (Twitter)

Education News & Politics Science & Technology

Anya Rossi• Live Now

Private livecam show

0 Comments

No comments available

Comments from the original post will appear here

Related Videos

Treating every inference request the same is lighting money on fire. DigitalOcean Inference Router sends each request to the right model automatically–lower cost, faster response, and no quality tradeoff. Open-source engine. Drop-in API. Live since April. ✨

Treating every inference request the same is lighting money on fire. DigitalOcean Inference Router sends each request to the right model automatically–lower cost, faster response, and no quality tradeoff. Open-source engine. Drop-in API. Live since April. ✨

DigitalOcean

594,950 views • 2 months ago

World models are moving beyond offline generation towards interactive, real-time experiences. Introducing ⚡FlashDreams⚡: an open-source high-performance inference and serving library built for autoregressive world models: 🔥 Up to 3.10× faster LingBot-World inference 🔥 Up to 2.12× faster Self-Forcing inference 🔥 Up to 1.40× faster Wan2.1 inference 🔥 8 integrated models 🔥 Multi-GPU, streaming, low-latency serving 🔥 Agentic skills that teach you how to use it FlashDreams is designed for a new generation of AI systems that continuously evolve over time while responding to user interactions. It powers applications across robotics, autonomous vehicle simulation, gaming, and virtual worlds. Github: Docs: Research page: Join the #flashdreams Discord channel at FlashDreams is also the runtime backbone behind NVIDIA OmniDreams ( 1/n #AI #WorldModels #FastInference #PhysicalAI #OpenSource #NVIDIA

World models are moving beyond offline generation towards interactive, real-time experiences. Introducing ⚡FlashDreams⚡: an open-source high-performance inference and serving library built for autoregressive world models: 🔥 Up to 3.10× faster LingBot-World inference 🔥 Up to 2.12× faster Self-Forcing inference 🔥 Up to 1.40× faster Wan2.1 inference 🔥 8 integrated models 🔥 Multi-GPU, streaming, low-latency serving 🔥 Agentic skills that teach you how to use it FlashDreams is designed for a new generation of AI systems that continuously evolve over time while responding to user interactions. It powers applications across robotics, autonomous vehicle simulation, gaming, and virtual worlds. Github: Docs: Research page: Join the #flashdreams Discord channel at FlashDreams is also the runtime backbone behind NVIDIA OmniDreams ( 1/n #AI #WorldModels #FastInference #PhysicalAI #OpenSource #NVIDIA

Ruilong Li

91,577 views • 2 months ago

locate-anything.cpp: native C++/ggml (ggml) inference for NVIDIA's LocateAnything-3B, open-vocabulary object detection / visual grounding, one of the neat detection VLMs out there, from the LocalAI team. Same detections as the official model, now running anywhere with no Python and faster, on CPU and GPU. Quantized GGUF on Hugging Face 🤗 Huge thanks to Georgi Gerganov for ggml and to NVIDIAAI for releasing LocateAnything! 🧵

locate-anything.cpp: native C++/ggml (ggml) inference for NVIDIA's LocateAnything-3B, open-vocabulary object detection / visual grounding, one of the neat detection VLMs out there, from the LocalAI team. Same detections as the official model, now running anywhere with no Python and faster, on CPU and GPU. Quantized GGUF on Hugging Face 🤗 Huge thanks to Georgi Gerganov for ggml and to NVIDIAAI for releasing LocateAnything! 🧵

Ettore Di Giacinto

13,972 views • 2 months ago

Excited to announce 🚀gsplat v1.0🚀: a ⏩efficient⏩ CUDA backend for 3D Gaussian Splatting! A drop-in replacement of the official impl. with: - Up to 2x faster training; - Up to 4x less GPU memory; - Render millions of GSs in real-time; - And more;

Excited to announce 🚀gsplat v1.0🚀: a ⏩efficient⏩ CUDA backend for 3D Gaussian Splatting! A drop-in replacement of the official impl. with: - Up to 2x faster training; - Up to 4x less GPU memory; - Render millions of GSs in real-time; - And more;

Ruilong Li

88,059 views • 2 years ago

The latest MLX has a CUDA back-end! To get started: pip install "mlx[cuda]" With the same codebase you can develop locally, run your model on Apple silicon, or in the cloud on Nvidia GPUs. MLX is designed around Apple silicon - which has a unified memory architecture. It uses the same design with CUDA. So there's no need to move arrays around from CPU memory to GPU memory. Note, this is early days - some operations are missing and performance is still being optimized. But it's already quite fast for Transformer training, text generation, and more! Here's a demo using mlx-lm to generate text with Llama 3 8B (bf16) on an A100:

The latest MLX has a CUDA back-end! To get started: pip install "mlx[cuda]" With the same codebase you can develop locally, run your model on Apple silicon, or in the cloud on Nvidia GPUs. MLX is designed around Apple silicon - which has a unified memory architecture. It uses the same design with CUDA. So there's no need to move arrays around from CPU memory to GPU memory. Note, this is early days - some operations are missing and performance is still being optimized. But it's already quite fast for Transformer training, text generation, and more! Here's a demo using mlx-lm to generate text with Llama 3 8B (bf16) on an A100:

Awni Hannun

42,761 views • 1 year ago

⚡️TRUMP MOVES FASTER THAN SILICON VALLEY Crypto Czar David Sacks says Trump is working at tech speed — even faster. 6 months in, historic wins are piling up. The U.S. is on track to become the "Crypto Capital of the World" 🇺🇸 How fast will adoption be? Faster than expected.

⚡️TRUMP MOVES FASTER THAN SILICON VALLEY Crypto Czar David Sacks says Trump is working at tech speed — even faster. 6 months in, historic wins are piling up. The U.S. is on track to become the "Crypto Capital of the World" 🇺🇸 How fast will adoption be? Faster than expected.

Coin Bureau

55,431 views • 1 year ago

ELON MUSK: "We're the first to achieve 100,000 H100 GPU training cluster, and we're now about to achieve the first 1 million H100 GPU equivalents in training. It's important to consider, for competitiveness of any technology company, what matters is not the position at any point in time, but what is your velocity and acceleration. And if you're moving faster than anyone else in any given technology arena, you will be the leader, and xAI is moving faster than any other company. No one's even close."

ELON MUSK: "We're the first to achieve 100,000 H100 GPU training cluster, and we're now about to achieve the first 1 million H100 GPU equivalents in training. It's important to consider, for competitiveness of any technology company, what matters is not the position at any point in time, but what is your velocity and acceleration. And if you're moving faster than anyone else in any given technology arena, you will be the leader, and xAI is moving faster than any other company. No one's even close."

DogeDesigner

37,810 views • 6 months ago

🗓️ Release Week Recap Big week. mlx-audio and mlx-vlm are now among some of the fastest-growing OSS projects. Here’s what we shipped last week. Gemma 4 on Apple Silicon Two awesome releases by our partner Google DeepMind & Google Gemma : > Gemma 4 12B — their new dense, unified multimodal model. It uses an encoder free audio path and simplified vision encoder. > Gemma 4 QAT — quantization-aware training checkpoints, optimized to run locally on consumer GPUs and edge devices, compressing the model while preserving the quality you expect from Gemma 4. On the audio 🎧 side we added support for 15+ new TTS, ASR & VAD models, faster long-form transcription, and an expanded OpenAI-compatible audio server. All local on Apple Silicon. Huge thanks to every contributor and my co-maintainer Lucas Newman. 🙏🏽

🗓️ Release Week Recap Big week. mlx-audio and mlx-vlm are now among some of the fastest-growing OSS projects. Here’s what we shipped last week. Gemma 4 on Apple Silicon Two awesome releases by our partner Google DeepMind & Google Gemma : > Gemma 4 12B — their new dense, unified multimodal model. It uses an encoder free audio path and simplified vision encoder. > Gemma 4 QAT — quantization-aware training checkpoints, optimized to run locally on consumer GPUs and edge devices, compressing the model while preserving the quality you expect from Gemma 4. On the audio 🎧 side we added support for 15+ new TTS, ASR & VAD models, faster long-form transcription, and an expanded OpenAI-compatible audio server. All local on Apple Silicon. Huge thanks to every contributor and my co-maintainer Lucas Newman. 🙏🏽

Prince Canuma

11,731 views • 2 months ago

Today we're shipping Nemotron 3 Ultra. A 550B MoE frontier-intelligence open model built for long-running agents. It delivers 5x faster inference and lowers the cost of complex agentic tasks by up to 30% versus other open frontier models.

Today we're shipping Nemotron 3 Ultra. A 550B MoE frontier-intelligence open model built for long-running agents. It delivers 5x faster inference and lowers the cost of complex agentic tasks by up to 30% versus other open frontier models.

NVIDIA AI

1,325,127 views • 2 months ago

At 6% the size of DeepSeek V4 Pro, Laguna S 2.1 delivers significantly higher performance. Today, we’re teaming up with Poolside to make Laguna XS 2.1 (same architecture as Laguna S 2.1) even faster on consumer Macs. We’re launching MLX(.)fast: a public competition where anyone can collaborate, contribute optimizations, and accelerate Laguna’s MLX inference engine.

At 6% the size of DeepSeek V4 Pro, Laguna S 2.1 delivers significantly higher performance. Today, we’re teaming up with Poolside to make Laguna XS 2.1 (same architecture as Laguna S 2.1) even faster on consumer Macs. We’re launching MLX(.)fast: a public competition where anyone can collaborate, contribute optimizations, and accelerate Laguna’s MLX inference engine.

Gajesh

106,400 views • 14 days ago

Wohoo!! 🎉 You can now generate 3D structures of proteins, RNA, DNA, and small molecules. 🚀 Check out BOLTZ! — the first open-source and commercially available model to achieve AlphaFold3-level accuracy in biomolecular structure prediction. 🧪 Try it yourself on Gradio ! Explore and generate different proteins and viruses: For faster inference, check it out here (on L4): 👉 Let's make biology more open on Hugging Face ! If there’s anything to fear, it’s tech being closed :) !

Wohoo!! 🎉 You can now generate 3D structures of proteins, RNA, DNA, and small molecules. 🚀 Check out BOLTZ! — the first open-source and commercially available model to achieve AlphaFold3-level accuracy in biomolecular structure prediction. 🧪 Try it yourself on Gradio ! Explore and generate different proteins and viruses: For faster inference, check it out here (on L4): 👉 Let's make biology more open on Hugging Face ! If there’s anything to fear, it’s tech being closed :) !

Jade Choghari

55,770 views • 1 year ago

Smart Turn v2: open source, native audio turn detection in 14 languages. New checkpoint of the open source, open data, open training code, semantic VAD model on Hugging Face, fal, and Pipecat AI. - 3x faster inference (12ms on an L40) - 14 languages (13 more than v1, which was english-only) - New synthetic data set `chirp_3_all` with ~163k audio samples - 99% accuracy on held out `human_5_all` test data Good turn detection is critical for voice agents. This model "understands" both semantic and audio patterns, and mitigates the voice AI trade-off between unwanted turn latency vs the agent interrupting people before they are finished speaking. Training scripts for both Modal and local training are in the repo. We want to make it as easy as possible to contribute to or customize this model! Here's a demo running the smart-turn model with default settings, aimed at generally hitting 400ms total turn detection time. You can tune things to be faster, too. You can help by contributing data, doing architecture expermints, or cleaning open source data! Keep reading ...

Smart Turn v2: open source, native audio turn detection in 14 languages. New checkpoint of the open source, open data, open training code, semantic VAD model on Hugging Face, fal, and Pipecat AI. - 3x faster inference (12ms on an L40) - 14 languages (13 more than v1, which was english-only) - New synthetic data set `chirp_3_all` with ~163k audio samples - 99% accuracy on held out `human_5_all` test data Good turn detection is critical for voice agents. This model "understands" both semantic and audio patterns, and mitigates the voice AI trade-off between unwanted turn latency vs the agent interrupting people before they are finished speaking. Training scripts for both Modal and local training are in the repo. We want to make it as easy as possible to contribute to or customize this model! Here's a demo running the smart-turn model with default settings, aimed at generally hitting 400ms total turn detection time. You can tune things to be faster, too. You can help by contributing data, doing architecture expermints, or cleaning open source data! Keep reading ...

kwindla

42,246 views • 1 year ago

Most people should probably update their priors on the state of open-source speech-to-speech. It's honestly kind of mind-blowing. We teamed up with Cerebras to build a fully open-source realtime voice demo (models + code) to show what's possible today. Demo : Blog: Go test it, fork it, tweak it, and impress your friends. video is raw, no cut, no speed-up, first take

Most people should probably update their priors on the state of open-source speech-to-speech. It's honestly kind of mind-blowing. We teamed up with Cerebras to build a fully open-source realtime voice demo (models + code) to show what's possible today. Demo : Blog: Go test it, fork it, tweak it, and impress your friends. video is raw, no cut, no speed-up, first take

Thomas Wolf

250,714 views • 1 month ago

Laguna S 2.1 is built to run locally. We heard the ask for a serious open model in the 80–120B range. Laguna S 2.1 is our answer, and we want it to redefine what this size class can do. Weights ship in BF16, FP8, INT4 and NVFP4, with official GGUF and MLX conversions, plus official DFlash draft models for faster inference. Day-one support across llama.cpp, ollama, Valerie Lee Make Up, SGLang, NVIDIA TRT-LLM, Hugging Face transformers and ZML. Watch it run on an NVIDIA AI DGX Spark.

Laguna S 2.1 is built to run locally. We heard the ask for a serious open model in the 80–120B range. Laguna S 2.1 is our answer, and we want it to redefine what this size class can do. Weights ship in BF16, FP8, INT4 and NVFP4, with official GGUF and MLX conversions, plus official DFlash draft models for faster inference. Day-one support across llama.cpp, ollama, Valerie Lee Make Up, SGLang, NVIDIA TRT-LLM, Hugging Face transformers and ZML. Watch it run on an NVIDIA AI DGX Spark.

Poolside

31,242 views • 21 days ago

Introducing Helmor The open-source, local-first answer to Conductor. A more refined, faster GUI for orchestrating coding agents. No cloud. One-click import from Conductor. AI made coding faster. Helmor is about finishing the rest of the loop: orchestration, workspaces, review, testing, and merge. We believe the next generation of GUI agent orchestration should be built in the open — by the community.

Introducing Helmor The open-source, local-first answer to Conductor. A more refined, faster GUI for orchestrating coding agents. No cloud. One-click import from Conductor. AI made coding faster. Helmor is about finishing the rest of the loop: orchestration, workspaces, review, testing, and merge. We believe the next generation of GUI agent orchestration should be built in the open — by the community.

Caspian 東澔

121,766 views • 3 months ago

Real-time inference is a big challenge for VLAs. We’ve been working on a way to amortize inference delays in π0.5. Our new Real-Time Chunking (RTC) method speeds up π0.5 by allowing the robot to “think” while it’s moving, which makes it quite a bit faster! 🧵👇

Real-time inference is a big challenge for VLAs. We’ve been working on a way to amortize inference delays in π0.5. Our new Real-Time Chunking (RTC) method speeds up π0.5 by allowing the robot to “think” while it’s moving, which makes it quite a bit faster! 🧵👇

Sergey Levine

73,688 views • 1 year ago

Scaling costs can be difficult. For a non-profit, it can be almost impossible. That's why One Acre Fund went all in on open source and used GitHub Copilot to speed up development—getting real-world help to farmers, faster. Here's their story. ⬇️

Scaling costs can be difficult. For a non-profit, it can be almost impossible. That's why One Acre Fund went all in on open source and used GitHub Copilot to speed up development—getting real-world help to farmers, faster. Here's their story. ⬇️

GitHub

25,564 views • 8 months ago

SPZ 4 is here. 🚀 3–5x faster encoding. 2x faster loading. 20x faster in the browser. Tens of millions of points, no sweat. The "JPG for 3D Gaussian splats" just leveled up. Big thanks to everyone who made this release happen, especially the team at Adobe. Read more: #NianticSpatial #Scaniverse #SPZ #GaussianSplat #Reconstruction #PhysicalAI #GeospatialAI #AI #Adobe

SPZ 4 is here. 🚀 3–5x faster encoding. 2x faster loading. 20x faster in the browser. Tens of millions of points, no sweat. The "JPG for 3D Gaussian splats" just leveled up. Big thanks to everyone who made this release happen, especially the team at Adobe. Read more: #NianticSpatial #Scaniverse #SPZ #GaussianSplat #Reconstruction #PhysicalAI #GeospatialAI #AI #Adobe

Niantic Spatial 🌎

63,528 views • 3 months ago

🚀 As we gear up for our Official Listing, on 18th March. Here’s a quick reminder of what DeepLink is all about! ⚡ Cutting-edge tech that eliminates lag by bringing GPU power closer to you. Get ready for a faster, smoother, and more powerful experience! 💡🔥 #DeepLink #GPU #TechInnovation #Listingsoon

🚀 As we gear up for our Official Listing, on 18th March. Here’s a quick reminder of what DeepLink is all about! ⚡ Cutting-edge tech that eliminates lag by bringing GPU power closer to you. Get ready for a faster, smoother, and more powerful experience! 💡🔥 #DeepLink #GPU #TechInnovation #Listingsoon

DeepLink

147,007 views • 1 year ago

Total War: ROME II - Emperor Edition returns to macOS on May 29th! This Apple Silicon-native port makes the game playable on modern Macs for the first time. Owners of ROME II on Steam can upgrade at no extra cost! Learn more:

Total War: ROME II - Emperor Edition returns to macOS on May 29th! This Apple Silicon-native port makes the game playable on modern Macs for the first time. Owners of ROME II on Steam can upgrade at no extra cost! Learn more:

Feral Interactive

27,010 views • 1 year ago