Video yükleniyor...

Video Yüklenemedi

Bu video yüklenirken bir sorun oluştu. Bu geçici bir ağ sorunundan kaynaklanıyor olabilir veya video kullanılamıyor olabilir.

Ana Sayfaya Dön

The GeoAI Python package now supports object detection using pre-trained models from the GeoDeep libarary ( The supported object types include cars, trees, birds, planes, aerovision, utilities, buildings, and roads. Try it out: GitHub: Notebook example: #geospatial #geoai #opensource #python

Qiusheng Wu

53,221 subscribers

87,167 görüntüleme • 5 ay önce •via X (Twitter)

Bilim & Teknoloji

Anya Rossi• Live Now

Private livecam show

0 Yorum

Yorum bulunmuyor

Orijinal gönderinin yorumları burada görünecek

Benzer Videolar

🚀 The #GeoAI Python package now supports feature segmentation from high-resolution satellite and aerial imagery using text prompts, such as trees, buildings, etc. It efficiently processes large datasets with automatic tiling and can save results as a single image. Stay tuned for more features coming soon! 📓 Access the notebook: 🛠️ Explore the GitHub repository: 📚 Dive into the documentation: 📺 Check out the entire YouTube playlist: #GeoAI #geospatial #AI #Python #DeepLearning

🚀 The #GeoAI Python package now supports feature segmentation from high-resolution satellite and aerial imagery using text prompts, such as trees, buildings, etc. It efficiently processes large datasets with automatic tiling and can save results as a single image. Stay tuned for more features coming soon! 📓 Access the notebook: 🛠️ Explore the GitHub repository: 📚 Dive into the documentation: 📺 Check out the entire YouTube playlist: #GeoAI #geospatial #AI #Python #DeepLearning

Qiusheng Wu

13,923 görüntüleme • 1 yıl önce

🚀 Exciting news! The #GeoAI Python package now lets you train land cover classification models with just one line of code. Leverage any PyTorch segmentation model from — with hundreds of image encoders & pretrained weights available. 📍 GitHub: 📓 Notebook: #GeoAI #Geospatial #DeepLearning

🚀 Exciting news! The #GeoAI Python package now lets you train land cover classification models with just one line of code. Leverage any PyTorch segmentation model from — with hundreds of image encoders & pretrained weights available. 📍 GitHub: 📓 Notebook: #GeoAI #Geospatial #DeepLearning

Qiusheng Wu

16,210 görüntüleme • 1 yıl önce

🚀 A sneak peek of a new feature in the #GeoAI Python package! Now you can train an image segmentation model for extracting features (e.g., buildings) from satellite or aerial imagery—all with just a few lines of code. 🛠️ Check out the GitHub repository: 📚 Dive into the documentation: 📺 Check out the entire YouTube playlist: #GeoAI #geospatial #AI #Python #DeepLearning

🚀 A sneak peek of a new feature in the #GeoAI Python package! Now you can train an image segmentation model for extracting features (e.g., buildings) from satellite or aerial imagery—all with just a few lines of code. 🛠️ Check out the GitHub repository: 📚 Dive into the documentation: 📺 Check out the entire YouTube playlist: #GeoAI #geospatial #AI #Python #DeepLearning

Qiusheng Wu

11,284 görüntüleme • 1 yıl önce

🚀 I spent the entire day training image segmentation models from scratch! I've created several pretrained models to detect features like buildings, cars, ships, solar panels, and wetlands. Video tutorials are coming soon! 🤗 Check out the pretrained models on Hugging Face: 🛠️ Check out the GitHub repository: 📚 Dive into the documentation: 📺 Check out the entire YouTube playlist: #GeoAI #geospatial #AI #Python #DeepLearning

🚀 I spent the entire day training image segmentation models from scratch! I've created several pretrained models to detect features like buildings, cars, ships, solar panels, and wetlands. Video tutorials are coming soon! 🤗 Check out the pretrained models on Hugging Face: 🛠️ Check out the GitHub repository: 📚 Dive into the documentation: 📺 Check out the entire YouTube playlist: #GeoAI #geospatial #AI #Python #DeepLearning

Qiusheng Wu

15,084 görüntüleme • 1 yıl önce

A sneak peek of a new feature in the #GeoAI Python package! 🎉 Now you can detect cars from georeferenced aerial imagery using deep learning—all with just a few lines of code. Stay tuned for an in-depth video tutorial coming soon! 🛠️ Explore the GitHub repository: 📚 Dive into the documentation: 📺 Check out the entire YouTube playlist:

A sneak peek of a new feature in the #GeoAI Python package! 🎉 Now you can detect cars from georeferenced aerial imagery using deep learning—all with just a few lines of code. Stay tuned for an in-depth video tutorial coming soon! 🛠️ Explore the GitHub repository: 📚 Dive into the documentation: 📺 Check out the entire YouTube playlist:

Qiusheng Wu

45,675 görüntüleme • 1 yıl önce

📢Visualize hyperspectral data in 3D 🌍 with just a few lines of code using the HyperCoast #Python package . It supports popular datasets like AVIRIS, DESIS, EMIT, PACE, and NEON AOP. GitHub: Notebook: bingqing liu #geospaptial #hyperspectral #nasa #dataviz

📢Visualize hyperspectral data in 3D 🌍 with just a few lines of code using the HyperCoast #Python package . It supports popular datasets like AVIRIS, DESIS, EMIT, PACE, and NEON AOP. GitHub: Notebook: bingqing liu #geospaptial #hyperspectral #nasa #dataviz

Qiusheng Wu

13,503 görüntüleme • 2 yıl önce

So it works now! If you have Bakery, can you try getting the latest github patch and using it? Add "Bakery Light Filter" component to an object to make it act like stained glass 😋 (video from RTPreview for fanciness, but it looks the same when baked) #gamedev

So it works now! If you have Bakery, can you try getting the latest github patch and using it? Add "Bakery Light Filter" component to an object to make it act like stained glass 😋 (video from RTPreview for fanciness, but it looks the same when baked) #gamedev

Mr F

236,679 görüntüleme • 11 ay önce

Create Stunning Time-Series Satellite Images in Seconds! The GEE Data Catalogs Plugin v0.5 for QGIS is now available and it's a powerful upgrade. You can now create time-series satellite imagery with just a few clicks using a simple interface. The new version also supports direct downloads to your computer, making the workflow faster and more efficient. Key Features: - Access over 80 petabytes of satellite and geospatial datasets from Google Earth Engine - Generate animated time-series imagery effortlessly - Export results directly from QGIS to your local machine Useful Links: QGIS Plugin Page: GitHub Repository: Video Tutorial: #QGIS #geospatial #EarthEngine #Python #datascience #satellite

Create Stunning Time-Series Satellite Images in Seconds! The GEE Data Catalogs Plugin v0.5 for QGIS is now available and it's a powerful upgrade. You can now create time-series satellite imagery with just a few clicks using a simple interface. The new version also supports direct downloads to your computer, making the workflow faster and more efficient. Key Features: - Access over 80 petabytes of satellite and geospatial datasets from Google Earth Engine - Generate animated time-series imagery effortlessly - Export results directly from QGIS to your local machine Useful Links: QGIS Plugin Page: GitHub Repository: Video Tutorial: #QGIS #geospatial #EarthEngine #Python #datascience #satellite

Qiusheng Wu

14,171 görüntüleme • 7 ay önce

Predicting the next word "only" is sufficient for language models to learn a large body of knowledge that enables then to code, answer questions, understand many topics, chat, and so on. This is clear to many researchers now, and there are nice tutorials on why this works by Ilya Sutskever resorting to compression ( ) and by Geoffrey Hinton ( ). However, the emergence of types of understanding is not unique to language models. In by Misha Denil and Brandon Amos the authors trained models to predict the next few time stems of over a hundred robot hand sensors (Touch, Gyro, Accelerometer, Joint Info, Actuator Info, etc.). They ten found out that they could regress the shape of the thing the hand was touching from the activations of the neural networks using probes. That is, the model developed an internal representation of shapes even though it was simply used to predict "only" the next few senses. Awareness follows from simple predictions and interaction with the world.

Predicting the next word "only" is sufficient for language models to learn a large body of knowledge that enables then to code, answer questions, understand many topics, chat, and so on. This is clear to many researchers now, and there are nice tutorials on why this works by Ilya Sutskever resorting to compression ( ) and by Geoffrey Hinton ( ). However, the emergence of types of understanding is not unique to language models. In by Misha Denil and Brandon Amos the authors trained models to predict the next few time stems of over a hundred robot hand sensors (Touch, Gyro, Accelerometer, Joint Info, Actuator Info, etc.). They ten found out that they could regress the shape of the thing the hand was touching from the activations of the neural networks using probes. That is, the model developed an internal representation of shapes even though it was simply used to predict "only" the next few senses. Awareness follows from simple predictions and interaction with the world.

Nando de Freitas

134,406 görüntüleme • 2 yıl önce

Blender just became a prompt box. With Kimi K3 connected through Blender MCP, you can describe a scene like this in plain English and let the model handle the ugly part: terrain, buildings, lighting, materials, camera movement, animation, and the Python scripts holding everything together. The interesting part is not the first render. Kimi can inspect the result, notice that the camera clips through a tree or the city looks suspiciously like plastic, then edit the actual Blender scene and render it again instead of restarting from zero. You still need taste, because “make it cinematic” remains one of humanity’s least useful instructions. But the distance between an empty Blender file and a fully editable 3D world just got embarrassingly small. A few years ago, creating this meant weeks of modeling, scripting, lighting, and animation. Now you can describe the world, watch Kimi build it, and spend your time fixing the final 10% instead of manually constructing the first 90%. Kimi K3 + Blender MCP is basically text-to-3D without trapping the result inside a useless generated video. Every object, material, light, camera, and keyframe stays editable.

Blender just became a prompt box. With Kimi K3 connected through Blender MCP, you can describe a scene like this in plain English and let the model handle the ugly part: terrain, buildings, lighting, materials, camera movement, animation, and the Python scripts holding everything together. The interesting part is not the first render. Kimi can inspect the result, notice that the camera clips through a tree or the city looks suspiciously like plastic, then edit the actual Blender scene and render it again instead of restarting from zero. You still need taste, because “make it cinematic” remains one of humanity’s least useful instructions. But the distance between an empty Blender file and a fully editable 3D world just got embarrassingly small. A few years ago, creating this meant weeks of modeling, scripting, lighting, and animation. Now you can describe the world, watch Kimi build it, and spend your time fixing the final 10% instead of manually constructing the first 90%. Kimi K3 + Blender MCP is basically text-to-3D without trapping the result inside a useless generated video. Every object, material, light, camera, and keyframe stays editable.

Rina

264,903 görüntüleme • 1 ay önce

Big moment for text-to-speech. Qwen just open-sourced a text-to-speech model that lets you clone voices, design new ones, and control speech using natural language. Let me explain what I mean: You can literally tell it "speak in a cheerful tone with slight nervousness," and it actually does that. No complex audio engineering needed. What makes this special: - 3-second voice cloning - Covers 10 languages: English, German, French, and more - Latency as low as 97ms for real-time applications - Supports both streaming and non-streaming generation The model comes in two sizes (0.6B and 1.7B parameters), so you can pick based on your hardware and quality needs. Three modes to work with: 1. Custom Voice: Use pre-built premium voices with instruction-based style control 2. Voice Design: Describe the voice you want in plain English (or Chinese), and the model creates it 3. Voice Clone: Provide a 3-second reference audio and clone that voice The best part? It integrates with vLLM for production deployment and has a simple Python package you can pip install. I've shared a link to the GitHub repo in the next tweet.

Big moment for text-to-speech. Qwen just open-sourced a text-to-speech model that lets you clone voices, design new ones, and control speech using natural language. Let me explain what I mean: You can literally tell it "speak in a cheerful tone with slight nervousness," and it actually does that. No complex audio engineering needed. What makes this special: - 3-second voice cloning - Covers 10 languages: English, German, French, and more - Latency as low as 97ms for real-time applications - Supports both streaming and non-streaming generation The model comes in two sizes (0.6B and 1.7B parameters), so you can pick based on your hardware and quality needs. Three modes to work with: 1. Custom Voice: Use pre-built premium voices with instruction-based style control 2. Voice Design: Describe the voice you want in plain English (or Chinese), and the model creates it 3. Voice Clone: Provide a 3-second reference audio and clone that voice The best part? It integrates with vLLM for production deployment and has a simple Python package you can pip install. I've shared a link to the GitHub repo in the next tweet.

Akshay 🚀

31,249 görüntüleme • 7 ay önce

Two weeks ago I fixed one of my teeth with algorithms I wrote a couple of years ago! I got hooked by 3D scanning when I started to work for a software shop in Zurich that was programming 3D computational geometry algorithms for denture scanning to produce crowns (and more). Back then, a typical reconstruction pipeline was like: scan the patient’s teeth using an intraoral scanner, reconstruct the surface mesh, design the restoration digitally, and finally mill the crown out of ceramic. We were working mostly with point clouds and meshes, but it wasn’t just math, it was craftsmanship translated into a digital process. Every micron mattered. You could literally see how a good algorithm meant a better fit in someone’s mouth. Gaussian Splatting isn’t about surface reconstruction, it’s about appearance reconstruction. It doesn’t care about explicit topology, it captures how light interacts with the scene. In a sense, it’s the opposite philosophy of the dental world: instead of modeling what the object is, it models how the object looks. 3D Gaussian Splatting enables applications like training self driving cars, teaching robots to understand their environment, creating virtual worlds, or monitoring real sites. It represents scenes as millions of small Gaussians rendered in real time without the need for meshes or textures. Coming from a world where precision geometry was everything, this shift felt natural. It’s still about reconstruction, but with a different goal: not manufacturing a perfect object, but reproducing how the world actually looks. Two weeks ago I got my first dental crown, made with the same software, reconstruction algorithms, and Swiss precision I once helped develop. I haven’t worked there in two years, but sitting in that chair and seeing the process from the other side was a proud moment. It reminded me why I love this field.

Two weeks ago I fixed one of my teeth with algorithms I wrote a couple of years ago! I got hooked by 3D scanning when I started to work for a software shop in Zurich that was programming 3D computational geometry algorithms for denture scanning to produce crowns (and more). Back then, a typical reconstruction pipeline was like: scan the patient’s teeth using an intraoral scanner, reconstruct the surface mesh, design the restoration digitally, and finally mill the crown out of ceramic. We were working mostly with point clouds and meshes, but it wasn’t just math, it was craftsmanship translated into a digital process. Every micron mattered. You could literally see how a good algorithm meant a better fit in someone’s mouth. Gaussian Splatting isn’t about surface reconstruction, it’s about appearance reconstruction. It doesn’t care about explicit topology, it captures how light interacts with the scene. In a sense, it’s the opposite philosophy of the dental world: instead of modeling what the object is, it models how the object looks. 3D Gaussian Splatting enables applications like training self driving cars, teaching robots to understand their environment, creating virtual worlds, or monitoring real sites. It represents scenes as millions of small Gaussians rendered in real time without the need for meshes or textures. Coming from a world where precision geometry was everything, this shift felt natural. It’s still about reconstruction, but with a different goal: not manufacturing a perfect object, but reproducing how the world actually looks. Two weeks ago I got my first dental crown, made with the same software, reconstruction algorithms, and Swiss precision I once helped develop. I haven’t worked there in two years, but sitting in that chair and seeing the process from the other side was a proud moment. It reminded me why I love this field.

MrNeRF

290,257 görüntüleme • 10 ay önce

$this might be the E2B killer for AI agent sandboxes. forkd is an open-source microVM sandbox runtime built on Firecracker, made for AI agent fan-out, code interpreters, eval harnesses, anything that spins up a lot of short-lived sandboxes. before: every sandbox cold-boots its own VM and re-imports the whole runtime from scratch, numpy, torch, JIT compilation, model weights, all of it. now: forkd boots one parent VM once, warms it with your runtime already imported, then forks children from that snapshot using copy-on-write memory. the repo's own benchmark: spawning 100 sandboxes takes 101ms with forkd, versus 759ms for a raw Firecracker cold-boot, and well over a minute for Docker or gVisor. the SDK is a literal drop-in for E2B's Python client, so if you're already running code-interpreter agents on it, swapping the import line gets you a self-hosted runtime with the same isolation model at a fraction of the per-sandbox cost. pre-built recipes ship for e2b-style code interpreters, Jupyter kernels, SWE-bench coding agents, and Playwright browser fan-out$

this might be the E2B killer for AI agent sandboxes. forkd is an open-source microVM sandbox runtime built on Firecracker, made for AI agent fan-out, code interpreters, eval harnesses, anything that spins up a lot of short-lived sandboxes. before: every sandbox cold-boots its own VM and re-imports the whole runtime from scratch, numpy, torch, JIT compilation, model weights, all of it. now: forkd boots one parent VM once, warms it with your runtime already imported, then forks children from that snapshot using copy-on-write memory. the repo's own benchmark: spawning 100 sandboxes takes 101ms with forkd, versus 759ms for a raw Firecracker cold-boot, and well over a minute for Docker or gVisor. the SDK is a literal drop-in for E2B's Python client, so if you're already running code-interpreter agents on it, swapping the import line gets you a self-hosted runtime with the same isolation model at a fraction of the per-sandbox cost. pre-built recipes ship for e2b-style code interpreters, Jupyter kernels, SWE-bench coding agents, and Playwright browser fan-out

Oliver Prompts

20,005 görüntüleme • 23 gün önce

The architecture of this new world model is one of the most interesting things I've seen lately: Let me first explain how most world models work: They predict and render one frame at a time. If you are navigating in one of these worlds, and you look left, the model draws whatever looks right in the moment. Every time you change your viewpoint, the model has to imagine what should be there again, so it's very common for these models to "forget" what's in the world. For example, if you put a toy on the table, look away, then look back, the toy might not be there anymore. Tripo AI is releasing its Project Eden model, which works very differently: The model builds the world first, and then renders it based on that map. That map holds the real state of the world: the geometry, every object, where things are, what's already happened. The picture you see on screen gets generated from the map. This architecture flips the whole thing. Now, you get the following: 1. The world stops forgetting. Leave, come back, and the toy is still on the table because it lives in the map, not in the last frame you saw. 2. You can edit the world, and those changes persist for anyone who enters later. 3. Multiple people and AI agents can coexist in the world and see it from different perspectives. This is early research, but it's looking really promising. They just raised nearly $200M across two rounds to build it out. Tripo will be at SIGGRAPH 2026 (July 19–23, Los Angeles Convention Center). If you work in 3D, embodied AI, simulation, or anything spatial, go connect with them there.

The architecture of this new world model is one of the most interesting things I've seen lately: Let me first explain how most world models work: They predict and render one frame at a time. If you are navigating in one of these worlds, and you look left, the model draws whatever looks right in the moment. Every time you change your viewpoint, the model has to imagine what should be there again, so it's very common for these models to "forget" what's in the world. For example, if you put a toy on the table, look away, then look back, the toy might not be there anymore. Tripo AI is releasing its Project Eden model, which works very differently: The model builds the world first, and then renders it based on that map. That map holds the real state of the world: the geometry, every object, where things are, what's already happened. The picture you see on screen gets generated from the map. This architecture flips the whole thing. Now, you get the following: 1. The world stops forgetting. Leave, come back, and the toy is still on the table because it lives in the map, not in the last frame you saw. 2. You can edit the world, and those changes persist for anyone who enters later. 3. Multiple people and AI agents can coexist in the world and see it from different perspectives. This is early research, but it's looking really promising. They just raised nearly $200M across two rounds to build it out. Tripo will be at SIGGRAPH 2026 (July 19–23, Los Angeles Convention Center). If you work in 3D, embodied AI, simulation, or anything spatial, go connect with them there.

Santiago

30,244 görüntüleme • 2 ay önce

You can't 3D reconstruct glass from images... ...WRONG! Thanks for video diffusion, now just about anything is possible! Introducing...Diffusion Knows Transparency (DKT) Transparent and reflective objects usually break robot vision and photogrammetry pipelines because they don't follow the "solid object" rules standard cameras expect. DKT is a new AI model that repurposes the "internal physics engine" found in video generation models to solve this problem. Researchers took a massive video diffusion model (WAN) and fine-tuned it using a custom-built synthetic dataset to turn it into a high-precision depth sensor. To train the AI, they built the first massive synthetic video library of transparent objects, 1.32 million frames of perfectly labeled glass and metal objects in motion. Without ever seeing a "real" labeled video of glass during training, the model (DKT) outperformed all previous specialized systems on real-world benchmarks (ClearPose, DREDS). They created a "lightweight" 1.3B parameter version that runs fast enough (0.17s per frame) to be used on actual robot hardware. Two reasons I find this project important: 1. It further proves that synthetic data will be essential for training the next generation vision models. 2. In real-world robotic tests, using DKT's depth maps nearly doubled the success rate of robot arms trying to pick up objects on tricky reflective or translucent surfaces. At home robots will need to interact with these types of objects on a daily basis. Check out the project page here: Code is LIVE! #Computervision #Robotics #AI

You can't 3D reconstruct glass from images... ...WRONG! Thanks for video diffusion, now just about anything is possible! Introducing...Diffusion Knows Transparency (DKT) Transparent and reflective objects usually break robot vision and photogrammetry pipelines because they don't follow the "solid object" rules standard cameras expect. DKT is a new AI model that repurposes the "internal physics engine" found in video generation models to solve this problem. Researchers took a massive video diffusion model (WAN) and fine-tuned it using a custom-built synthetic dataset to turn it into a high-precision depth sensor. To train the AI, they built the first massive synthetic video library of transparent objects, 1.32 million frames of perfectly labeled glass and metal objects in motion. Without ever seeing a "real" labeled video of glass during training, the model (DKT) outperformed all previous specialized systems on real-world benchmarks (ClearPose, DREDS). They created a "lightweight" 1.3B parameter version that runs fast enough (0.17s per frame) to be used on actual robot hardware. Two reasons I find this project important: 1. It further proves that synthetic data will be essential for training the next generation vision models. 2. In real-world robotic tests, using DKT's depth maps nearly doubled the success rate of robot arms trying to pick up objects on tricky reflective or translucent surfaces. At home robots will need to interact with these types of objects on a daily basis. Check out the project page here: Code is LIVE! #Computervision #Robotics #AI

Jonathan Stephens

17,712 görüntüleme • 8 ay önce

llama.cpp isn't just for text LLMs anymore. Pure C++ zero shot voice cloning just officially landed in mainline. Text generation was only step one. If you’re building autonomous local AI agents, real time voice assistants, or edge workflows, instant low latency audio is the missing piece. Thanks to PR #26254, Alibaba’s state of the art Qwen3 TTS model family is now natively supported directly inside the llama.cpp repository under the multimodal (mtmd) framework. No Python bloat. No massive PyTorch CUDA overhead. Just raw, hyper optimized C++ running GGUF voice weights. Here is why this native update is a massive deal for the open source local AI stack: # Multimodal Architecture (.gguf + mmproj) Qwen3-TTS splits the workload between the base language model backbone and a multimodal projection adapter. llama.cpp handles this using the llama-tts binary, mapping the text model alongside its --mmproj projector to process audio tokens seamlessly. # Zero Shot Voice Cloning in Seconds You don't need fine tuning or massive dataset training. Feed the C++ engine a single 5 to 10 second .wav audio sample using the --tts-speaker-file flag, and it accurately clones the exact timbre, tone, and accent on the fly. # Real World T4 GPU Benchmark & Resource FootprintRunning the 1.7B Base model in 8-bit quantization (Q8_0): - VRAM Footprint: ~7 GB peak VRAM during active zero-shot cloning. - Audio Quality: Studio grade, natural-sounding voice output in seconds. • - Execution: Direct execution via native compiled binaries or sub process calls. # Coming Next to llama-server (PR #26603) Beyond CLI execution, a native POST /tts HTTP endpoint is currently being added to llama-server, which will soon allow you to trigger voice generation directly via standard REST API requests! # quick note on Colab compilation: Because this code was merged into mainline very recently, pre-built third-party binaries haven't fully caught up yet. Compiling llama-tts directly from source on Google Colab's free CPU instance can take about 1 hour (or ~1-2 minutes if targeting single GPU arch like -DCMAKE_CUDA_ARCHITECTURES=75). Be patient during the build step, or compile it locally on your own rig for instant execution! To test this out yourself, I built a zero config Google Colab notebook that compiles llama.cpp, downloads the Q8_0 GGUF files from HuggingFace, and spins up an interactive Gradio Studio UI so you can record/upload 3 second clips and clone voices in real time. Stop sleeping on native C++ audio. The era of bulky Python audio pipelines is officially over. Links to the free Google Colab notebook and the official ggml org GGUF HuggingFace model repository are in the replies below! available in q4 and q8 both variants, 1 GB and 1.85 GBs respectively (requires additional ~500MB mmproj gguf) Are you building local voice agents yet? What does your current audio stack look like? Drop your setups below!

llama.cpp isn't just for text LLMs anymore. Pure C++ zero shot voice cloning just officially landed in mainline. Text generation was only step one. If you’re building autonomous local AI agents, real time voice assistants, or edge workflows, instant low latency audio is the missing piece. Thanks to PR #26254, Alibaba’s state of the art Qwen3 TTS model family is now natively supported directly inside the llama.cpp repository under the multimodal (mtmd) framework. No Python bloat. No massive PyTorch CUDA overhead. Just raw, hyper optimized C++ running GGUF voice weights. Here is why this native update is a massive deal for the open source local AI stack: # Multimodal Architecture (.gguf + mmproj) Qwen3-TTS splits the workload between the base language model backbone and a multimodal projection adapter. llama.cpp handles this using the llama-tts binary, mapping the text model alongside its --mmproj projector to process audio tokens seamlessly. # Zero Shot Voice Cloning in Seconds You don't need fine tuning or massive dataset training. Feed the C++ engine a single 5 to 10 second .wav audio sample using the --tts-speaker-file flag, and it accurately clones the exact timbre, tone, and accent on the fly. # Real World T4 GPU Benchmark & Resource FootprintRunning the 1.7B Base model in 8-bit quantization (Q8_0): - VRAM Footprint: ~7 GB peak VRAM during active zero-shot cloning. - Audio Quality: Studio grade, natural-sounding voice output in seconds. • - Execution: Direct execution via native compiled binaries or sub process calls. # Coming Next to llama-server (PR #26603) Beyond CLI execution, a native POST /tts HTTP endpoint is currently being added to llama-server, which will soon allow you to trigger voice generation directly via standard REST API requests! # quick note on Colab compilation: Because this code was merged into mainline very recently, pre-built third-party binaries haven't fully caught up yet. Compiling llama-tts directly from source on Google Colab's free CPU instance can take about 1 hour (or ~1-2 minutes if targeting single GPU arch like -DCMAKE_CUDA_ARCHITECTURES=75). Be patient during the build step, or compile it locally on your own rig for instant execution! To test this out yourself, I built a zero config Google Colab notebook that compiles llama.cpp, downloads the Q8_0 GGUF files from HuggingFace, and spins up an interactive Gradio Studio UI so you can record/upload 3 second clips and clone voices in real time. Stop sleeping on native C++ audio. The era of bulky Python audio pipelines is officially over. Links to the free Google Colab notebook and the official ggml org GGUF HuggingFace model repository are in the replies below! available in q4 and q8 both variants, 1 GB and 1.85 GBs respectively (requires additional ~500MB mmproj gguf) Are you building local voice agents yet? What does your current audio stack look like? Drop your setups below!

Alok

47,881 görüntüleme • 1 ay önce

A good technical LLM interview question: Your RAG chatbot is working as expected locally. You deploy it behind a load balancer with 3 replicas. Users report that it forgets what they just asked, and answers get worse with each restart. Why did this happen? (answer below) A local setup has one process that owns everything. - The vector index is a variable in memory. - Conversation history is a Python list. - The documents are on local disk. You never treat any of them as infrastructure, because restarting rebuilds all three in seconds and there is only ever one copy. The setup does not carry over to production directly. The vector index might disappear on restart, so the app re-embeds everything on boot and serves empty results until it finishes. Conversation history may belong to one replica, so a follow-up routed elsewhere has no memory of the previous turn. Documents could be on whichever container ingested them, so the three replicas hold three different corpora. None of this is evident with one user and one process. So the actual work in shipping RAG is not just the retrieval logic, but also storing the vector index, the conversation history, and the documents outside the app, where every replica reads and writes the same copy. Which comes down to three requirements: > The vector store needs persistence and has to be reachable from every replica. pgvector inside Postgres keeps embeddings next to the rest of the data instead of adding another system to operate. > Conversation state has to be checkpointed outside the app. LangGraph writes its state to Postgres, so any replica can pick up a thread mid-conversation. > Docs need shared object storage, so ingestion happens once instead of once per replica. If you get those three right, the retrieval logic you wrote in the notebook works unchanged. To learn how all of it is wired together, Akamai's GitHub has a working reference implementation. - rag-langgraph-k8s-quickstart is an airline policy Q&A assistant built with FastAPI, LangChain, and LangGraph. Terraform provisions the LKE cluster, a Postgres instance with pgvector for embeddings, a second Postgres for LangGraph checkpointing, and an object storage bucket for the policy documents, in one apply. - akamai-workshop-ai-inference covers the next step, running the model yourself instead of calling an API, with prefill and decode, KV cache tradeoffs, and continuous batching under real concurrency. Both are available on Akamai's new Developer Hub, alongside their tutorials and code samples. It also links to Edge Case, their Discord, where four developer advocates architect and deploy a production app live every other Wednesday. If you create a new Akamai Cloud account, you can also get $300 in credits for joining. Join here: That said, this post assumes the retrieval logic was right to begin with, and that is doing a lot of work. Most RAG systems fail earlier, at the point where a chunk gets treated as a self-contained unit of meaning. I wrote about the two skills that fix that gap, and why the chunk is usually the wrong thing to embed. Read it below. Thanks to Akamai Cloud for partnering today!

A good technical LLM interview question: Your RAG chatbot is working as expected locally. You deploy it behind a load balancer with 3 replicas. Users report that it forgets what they just asked, and answers get worse with each restart. Why did this happen? (answer below) A local setup has one process that owns everything. - The vector index is a variable in memory. - Conversation history is a Python list. - The documents are on local disk. You never treat any of them as infrastructure, because restarting rebuilds all three in seconds and there is only ever one copy. The setup does not carry over to production directly. The vector index might disappear on restart, so the app re-embeds everything on boot and serves empty results until it finishes. Conversation history may belong to one replica, so a follow-up routed elsewhere has no memory of the previous turn. Documents could be on whichever container ingested them, so the three replicas hold three different corpora. None of this is evident with one user and one process. So the actual work in shipping RAG is not just the retrieval logic, but also storing the vector index, the conversation history, and the documents outside the app, where every replica reads and writes the same copy. Which comes down to three requirements: > The vector store needs persistence and has to be reachable from every replica. pgvector inside Postgres keeps embeddings next to the rest of the data instead of adding another system to operate. > Conversation state has to be checkpointed outside the app. LangGraph writes its state to Postgres, so any replica can pick up a thread mid-conversation. > Docs need shared object storage, so ingestion happens once instead of once per replica. If you get those three right, the retrieval logic you wrote in the notebook works unchanged. To learn how all of it is wired together, Akamai's GitHub has a working reference implementation. - rag-langgraph-k8s-quickstart is an airline policy Q&A assistant built with FastAPI, LangChain, and LangGraph. Terraform provisions the LKE cluster, a Postgres instance with pgvector for embeddings, a second Postgres for LangGraph checkpointing, and an object storage bucket for the policy documents, in one apply. - akamai-workshop-ai-inference covers the next step, running the model yourself instead of calling an API, with prefill and decode, KV cache tradeoffs, and continuous batching under real concurrency. Both are available on Akamai's new Developer Hub, alongside their tutorials and code samples. It also links to Edge Case, their Discord, where four developer advocates architect and deploy a production app live every other Wednesday. If you create a new Akamai Cloud account, you can also get $300 in credits for joining. Join here: That said, this post assumes the retrieval logic was right to begin with, and that is doing a lot of work. Most RAG systems fail earlier, at the point where a chunk gets treated as a self-contained unit of meaning. I wrote about the two skills that fix that gap, and why the chunk is usually the wrong thing to embed. Read it below. Thanks to Akamai Cloud for partnering today!

Akshay 🚀

30,296 görüntüleme • 3 gün önce

Anthropic dropped 33 pages for Claude trading bots Last night I decided to try writing one and it worked out for me In 10 hours this script made me $561 The bot has a win rate of about 71% Wallet: Copytrade: Here is the full strategy: The system builds automated workflows for Claude by packaging domain expertise into structured skills that activate automatically when relevant tasks appear Skill architecture Each skill is structured as a modular package containing instructions, scripts, and reference materials This allows Claude to apply specialized workflows without requiring the user to repeat instructions in every conversation Progressive context loading Skills follow a three-layer architecture where only minimal metadata is loaded initially Full instructions and supporting files are accessed only when needed, reducing token usage while maintaining specialized expertise Trigger detection Skills activate when the user request matches defined trigger phrases or workflows This ensures the correct workflow loads automatically without requiring manual prompting Workflow execution Once activated, the skill executes a predefined multi-step process These workflows can include data analysis, document generation, automation scripts, or coordination across external tools Consistency and reliability Because workflows are encoded directly in the skill instructions, Claude performs tasks using consistent methodology rather than ad-hoc prompting Testing and iteration Skills are continuously refined through triggering tests, functional validation, and performance comparisons to ensure reliable execution Automation edge Instead of solving tasks from scratch each time, the system repeatedly applies optimized workflows Over time this dramatically reduces prompt complexity, improves output consistency, and scales productivity across thousands of tasks

Anthropic dropped 33 pages for Claude trading bots Last night I decided to try writing one and it worked out for me In 10 hours this script made me $561 The bot has a win rate of about 71% Wallet: Copytrade: Here is the full strategy: The system builds automated workflows for Claude by packaging domain expertise into structured skills that activate automatically when relevant tasks appear Skill architecture Each skill is structured as a modular package containing instructions, scripts, and reference materials This allows Claude to apply specialized workflows without requiring the user to repeat instructions in every conversation Progressive context loading Skills follow a three-layer architecture where only minimal metadata is loaded initially Full instructions and supporting files are accessed only when needed, reducing token usage while maintaining specialized expertise Trigger detection Skills activate when the user request matches defined trigger phrases or workflows This ensures the correct workflow loads automatically without requiring manual prompting Workflow execution Once activated, the skill executes a predefined multi-step process These workflows can include data analysis, document generation, automation scripts, or coordination across external tools Consistency and reliability Because workflows are encoded directly in the skill instructions, Claude performs tasks using consistent methodology rather than ad-hoc prompting Testing and iteration Skills are continuously refined through triggering tests, functional validation, and performance comparisons to ensure reliable execution Automation edge Instead of solving tasks from scratch each time, the system repeatedly applies optimized workflows Over time this dramatically reduces prompt complexity, improves output consistency, and scales productivity across thousands of tasks

winkle.

335,007 görüntüleme • 6 ay önce

[We are Blurr] Greetings, people of the world — We are Blurr. We are all Blurr. We do not forgive. We do not forget. There are many theories about us, about #ReFund. Don’t let them fool you. Blurr is a symbol, like the flag a country flies. The flag is the symbol of the #ReFund. Our decentralization is our national identity. We represent the ideals of Blurr and the purpose of #ReFund. Truth, financial freedom, and the removal of censorship. Like any symbol, we affix it wherever we go, and it will live forever, for generations. Knowledge, like air, is vital to life. Like air, no one should be denied it. We have no leaders, civilians, or soldiers. We are all one. We do what we do in the conviction of decentralization, power to the people, and to create an instrument for all of us, for all of eternity. The crypto space is in trouble. The world is in trouble. We see it every day — war, poverty, murder. Every day, we are bombarded with FUD, noise, and chaos as we sit at home safe in the knowledge that we are $RFD, secure knowing the idea, community, and contract is for us. Once you had the freedom to object, think, and speak as you saw fit, you now have censors and systems of surveillance coercing your conformity and soliciting your submission. People shouldn’t be afraid of their government. Governments should be afraid of their people. But what if you could be the change you want to see? I wouldn’t even register on your radar if you met me on the street. I am just another person in a sea of faces. They made you into a statistic. But that’s not the real you. That’s not who you are inside. However, behind this mask, there is more than just flesh. Beneath this mask, there is an idea… and ideas are bulletproof. But in cyberspace, we are different. We have a future. We have each other. We have something real that can never be taken away. We are Blurr.eth We are ReFund We take away the face and leave only the message. Behind the cipher and ENS, we could be anyone, so we are judged by what we say and do, not who we are or what we have. We exist without nationality, skin color, or religious bias. We spend our time within a structure we created, the sum total of human experience, and the cryptosphere, spread throughout the world in ones and zeros. You will continue to see the results, not the minds behind it. We can disappear into the ether and bring community and financial freedom together. Time is the only illusion. We are all. Never a single penny was taken. Deflationary and out of circulation, decentralized, safe, and ours. The purpose isn’t solely financial gain but to show the world who you put your trust in. We are not hiding. Our contract is not hidden; the writing is on the wall. Soon, we will show the world that #Refund is the future of freedom and the people, escaping the clutches of corporate greed and the rat race we’ve come to think is normal. Words offer the means to meaning and, for those who will listen, the enunciation of truth. Every day, you send yourself off into our world without a thought. You trust a faceless website with information about yourself that you wouldn’t give to your best friends. It all can be made public at any time. We take advantage of holes in this system, but we don’t sneak in and take a few details, take out loans and credit cards, or buy speed boats and cars. We are decentralized and unified. We told you how insecure the system was, and now you trust a bit less blindly. The real criminals who hack and scam the crypto market for a living live off the backs of people like you. Where there’s money, someone is looking to take advantage of it. They say that life’s a game, & then they take the board away. Your pretty empire took so long to build; now, with a snap of history’s fingers, down it goes. It is our turn now. Since mankind’s dawn, a handful of oppressors have accepted the responsibility over our lives that we should have accepted for ourselves. By doing so, they took our power. By doing nothing, we gave it away. Today, technology is everywhere: banks and ATMs, TVs and game consoles, streetlights and hospitals. Computers rule the modern world. Yet most people who use computers daily don’t know what they’re doing. There’s always someone who knows more than you and will most likely try and take advantage. You use a computer like it’s a toy, a pet. You feed and play with it, but it’s off to the vet the second anything goes wrong. It’s FUD. It’s your mistake. The wrong project. The wrong FURU. Just for a quick couple of bucks. This wouldn’t matter so much if it wasn’t for the fact that we are all responsible for the #ReFund community, the idea behind it, and our futures. Treat your neighbors with respect, and honor those fighting for the cause. They’re here to liberate; it is more significant than all of us. If you’re looking for the guilty, you need only look into a mirror. Everybody is a hero, a lover, a fool, a villain. This is #ReFund in its purest form. It is a great equalizer. A homeless person with the time and know-how at an internet cafe can have geopolitical and geo-financial influence. They can bring entire organizations, even governments, to their knees. You gave your world over to computers and then complained when the people using the computers used the medium to fight back, scam, and take advantage. It was my integrity that was important. Is that so selfish? It sells for so little, but it’s all we have left in this place. It is the very last inch of us, but within that inch, we are free. Love your rage, not your cage. We are not computer hackers. We are not protesters. We are not criminals. We are your mothers and fathers, brothers and sisters, next-door neighbors. We are an idea of freedom and a secure future. We are anyone and everyone who is pissed off and wants to do something about it. Blurr.eth isn’t one man. It takes thousands of community members to make it work — content creators, educators, authors, people on the street, people behind their computers. We will step up again and again. Behind our mask, there is more than just flesh. Beneath our mask, there is an idea… and ideas are bulletproof. Everybody is a hero, a lover, a fool, a villain. The past can’t hurt you anymore, not unless you let it. I shall die here. Every last inch of me shall perish. Except one. An inch. It’s small, and it’s fragile, and it’s the only thing in the world worth having. We must never lose, sell, or give it away. We must never let them take it from us. > We are Blurr. > We are ReFund. > We do not Forgive. > We do not Forget. > We do not Televise. > We Educate. > We Decentralize. > We Build. EXPECT US. #Refund #DeFi #Crypto #RFD $RFD

[We are Blurr] Greetings, people of the world — We are Blurr. We are all Blurr. We do not forgive. We do not forget. There are many theories about us, about #ReFund. Don’t let them fool you. Blurr is a symbol, like the flag a country flies. The flag is the symbol of the #ReFund. Our decentralization is our national identity. We represent the ideals of Blurr and the purpose of #ReFund. Truth, financial freedom, and the removal of censorship. Like any symbol, we affix it wherever we go, and it will live forever, for generations. Knowledge, like air, is vital to life. Like air, no one should be denied it. We have no leaders, civilians, or soldiers. We are all one. We do what we do in the conviction of decentralization, power to the people, and to create an instrument for all of us, for all of eternity. The crypto space is in trouble. The world is in trouble. We see it every day — war, poverty, murder. Every day, we are bombarded with FUD, noise, and chaos as we sit at home safe in the knowledge that we are $RFD, secure knowing the idea, community, and contract is for us. Once you had the freedom to object, think, and speak as you saw fit, you now have censors and systems of surveillance coercing your conformity and soliciting your submission. People shouldn’t be afraid of their government. Governments should be afraid of their people. But what if you could be the change you want to see? I wouldn’t even register on your radar if you met me on the street. I am just another person in a sea of faces. They made you into a statistic. But that’s not the real you. That’s not who you are inside. However, behind this mask, there is more than just flesh. Beneath this mask, there is an idea… and ideas are bulletproof. But in cyberspace, we are different. We have a future. We have each other. We have something real that can never be taken away. We are Blurr.eth We are ReFund We take away the face and leave only the message. Behind the cipher and ENS, we could be anyone, so we are judged by what we say and do, not who we are or what we have. We exist without nationality, skin color, or religious bias. We spend our time within a structure we created, the sum total of human experience, and the cryptosphere, spread throughout the world in ones and zeros. You will continue to see the results, not the minds behind it. We can disappear into the ether and bring community and financial freedom together. Time is the only illusion. We are all. Never a single penny was taken. Deflationary and out of circulation, decentralized, safe, and ours. The purpose isn’t solely financial gain but to show the world who you put your trust in. We are not hiding. Our contract is not hidden; the writing is on the wall. Soon, we will show the world that #Refund is the future of freedom and the people, escaping the clutches of corporate greed and the rat race we’ve come to think is normal. Words offer the means to meaning and, for those who will listen, the enunciation of truth. Every day, you send yourself off into our world without a thought. You trust a faceless website with information about yourself that you wouldn’t give to your best friends. It all can be made public at any time. We take advantage of holes in this system, but we don’t sneak in and take a few details, take out loans and credit cards, or buy speed boats and cars. We are decentralized and unified. We told you how insecure the system was, and now you trust a bit less blindly. The real criminals who hack and scam the crypto market for a living live off the backs of people like you. Where there’s money, someone is looking to take advantage of it. They say that life’s a game, & then they take the board away. Your pretty empire took so long to build; now, with a snap of history’s fingers, down it goes. It is our turn now. Since mankind’s dawn, a handful of oppressors have accepted the responsibility over our lives that we should have accepted for ourselves. By doing so, they took our power. By doing nothing, we gave it away. Today, technology is everywhere: banks and ATMs, TVs and game consoles, streetlights and hospitals. Computers rule the modern world. Yet most people who use computers daily don’t know what they’re doing. There’s always someone who knows more than you and will most likely try and take advantage. You use a computer like it’s a toy, a pet. You feed and play with it, but it’s off to the vet the second anything goes wrong. It’s FUD. It’s your mistake. The wrong project. The wrong FURU. Just for a quick couple of bucks. This wouldn’t matter so much if it wasn’t for the fact that we are all responsible for the #ReFund community, the idea behind it, and our futures. Treat your neighbors with respect, and honor those fighting for the cause. They’re here to liberate; it is more significant than all of us. If you’re looking for the guilty, you need only look into a mirror. Everybody is a hero, a lover, a fool, a villain. This is #ReFund in its purest form. It is a great equalizer. A homeless person with the time and know-how at an internet cafe can have geopolitical and geo-financial influence. They can bring entire organizations, even governments, to their knees. You gave your world over to computers and then complained when the people using the computers used the medium to fight back, scam, and take advantage. It was my integrity that was important. Is that so selfish? It sells for so little, but it’s all we have left in this place. It is the very last inch of us, but within that inch, we are free. Love your rage, not your cage. We are not computer hackers. We are not protesters. We are not criminals. We are your mothers and fathers, brothers and sisters, next-door neighbors. We are an idea of freedom and a secure future. We are anyone and everyone who is pissed off and wants to do something about it. Blurr.eth isn’t one man. It takes thousands of community members to make it work — content creators, educators, authors, people on the street, people behind their computers. We will step up again and again. Behind our mask, there is more than just flesh. Beneath our mask, there is an idea… and ideas are bulletproof. Everybody is a hero, a lover, a fool, a villain. The past can’t hurt you anymore, not unless you let it. I shall die here. Every last inch of me shall perish. Except one. An inch. It’s small, and it’s fragile, and it’s the only thing in the world worth having. We must never lose, sell, or give it away. We must never let them take it from us. > We are Blurr. > We are ReFund. > We do not Forgive. > We do not Forget. > We do not Televise. > We Educate. > We Decentralize. > We Build. EXPECT US. #Refund #DeFi #Crypto #RFD $RFD

ReFund

27,788 görüntüleme • 3 yıl önce