正在加载视频...

视频加载失败

加载此视频时出现问题。这可能是由于临时网络问题，或视频可能不可用。

🤗 MOSS-Transcribe-Diarize-0.9B is now open source on Hugging Face. Built with an end-to-end audio-to-structured-transcript paradigm: >0.9B open-source ASR model >Apache license 2.0 >128k long-context transcription >Up to ~90-min audio input >Speaker labels + timestamps in one generation >Multi-speaker diarization for meetings, interruptions, and overlapping voices >Hotword biasing for names,... show more

MOSI

1,080 subscribers

150,135 次观看 • 6 天前 •via X (Twitter)

科学技术教育

Anya Rossi• Live Now

Private livecam show

0 条评论

暂无评论

原始帖子的评论将显示在这里

相关视频

🤗 MOSS-VL-Realtime is now open source on Hugging Face . Built for real-time visual understanding over continuous video streams: 🧠 11B vision-language model 📜 Apache-2.0 license 💬 Ask questions at any point in a video stream 👀 Keeps watching while generating a response 🔄 Revises or interrupts its response as the scene changes 🤫 Can remain silent when more evidence is needed 🧩 256K-token context window 🌐 Chinese and English multimodal understanding 📦 Base, Instruct, and Realtime models are all open source From “watch first, answer later” to “keep watching while answering.” 👀 OpenMOSS Thank you SGLang LMSYS Org for day-0 support! 🚀

🤗 MOSS-VL-Realtime is now open source on Hugging Face . Built for real-time visual understanding over continuous video streams: 🧠 11B vision-language model 📜 Apache-2.0 license 💬 Ask questions at any point in a video stream 👀 Keeps watching while generating a response 🔄 Revises or interrupts its response as the scene changes 🤫 Can remain silent when more evidence is needed 🧩 256K-token context window 🌐 Chinese and English multimodal understanding 📦 Base, Instruct, and Realtime models are all open source From “watch first, answer later” to “keep watching while answering.” 👀 OpenMOSS Thank you SGLang LMSYS Org for day-0 support! 🚀

MOSI

42,421 次观看 • 1 天前

Microsoft did it again! Speech AI models have a major limitation. They slice long recordings into tiny chunks, lose track of who's speaking, and forget all context halfway through. This is exactly what Microsoft's VibeVoice solves. It's an open-source family of frontier voice AI models for both speech recognition and speech generation. Here's what it can do: > VibeVoice-ASR processes up to 60 minutes of audio in a single pass. No chunking. It outputs structured transcriptions with who spoke, when they spoke, and what they said. > You can feed it custom hotwords like names, technical jargon, or domain-specific terms. The model uses them to significantly improve accuracy on specialized content. > VibeVoice-TTS generates up to 90 minutes of multi-speaker speech with up to 4 distinct speakers. Natural turn-taking, emotional expression, all in one pass. > VibeVoice-Realtime is a 0.5B streaming TTS model with ~300ms first-audio latency. Small enough to deploy practically anywhere. All of this is powered by continuous speech tokenizers running at just 7.5 Hz. This ultra-low frame rate preserves audio quality while making long sequences computationally feasible. I have shared the link to the GitHub repo in the replies!

Microsoft did it again! Speech AI models have a major limitation. They slice long recordings into tiny chunks, lose track of who's speaking, and forget all context halfway through. This is exactly what Microsoft's VibeVoice solves. It's an open-source family of frontier voice AI models for both speech recognition and speech generation. Here's what it can do: > VibeVoice-ASR processes up to 60 minutes of audio in a single pass. No chunking. It outputs structured transcriptions with who spoke, when they spoke, and what they said. > You can feed it custom hotwords like names, technical jargon, or domain-specific terms. The model uses them to significantly improve accuracy on specialized content. > VibeVoice-TTS generates up to 90 minutes of multi-speaker speech with up to 4 distinct speakers. Natural turn-taking, emotional expression, all in one pass. > VibeVoice-Realtime is a 0.5B streaming TTS model with ~300ms first-audio latency. Small enough to deploy practically anywhere. All of this is powered by continuous speech tokenizers running at just 7.5 Hz. This ultra-low frame rate preserves audio quality while making long sequences computationally feasible. I have shared the link to the GitHub repo in the replies!

Akshay 🚀

45,206 次观看 • 3 个月前

Today we're announcing the open-source release of HunyuanVideo-Foley, our new end-to-end Text-Video-to-Audio (TV2A) framework for generating high-fidelity audio.🚀 This tool empowers creators in video production, filmmaking, and game development to generate professional-grade audio that precisely aligns with visual dynamics and semantic context, addressing key challenges in V2A generation.🔊 Key Innovations: 🔹Exceptional Generalization: Trained on a massive 100k-hour multimodal dataset, the model generates contextually-aware soundscapes for a wide range of scenes, from natural landscapes to animated shorts. 🔹Balanced Multimodal Response: Our innovative multimodal diffusion transformer (MMDiT) architecture ensures the model balances video and text cues, generating rich, layered sound effects that capture every detail—from the main subject to subtle background elements. 🔹High-Fidelity Audio: Using a Representation Alignment (REPA) loss function and a powerful Audio VAE, we've improved generation stability and producing professional-grade audio, free of noise and inconsistencies. HunyuanVideo-Foley achieves SOTA on multiple benchmarks, surpassing all open-source models in audio quality, visual-semantic alignment, and temporal alignment. 👉Try it now: 🌐Project Page: 🔗Code: 📄Technical Report: 🤗Hugging Face:

Today we're announcing the open-source release of HunyuanVideo-Foley, our new end-to-end Text-Video-to-Audio (TV2A) framework for generating high-fidelity audio.🚀 This tool empowers creators in video production, filmmaking, and game development to generate professional-grade audio that precisely aligns with visual dynamics and semantic context, addressing key challenges in V2A generation.🔊 Key Innovations: 🔹Exceptional Generalization: Trained on a massive 100k-hour multimodal dataset, the model generates contextually-aware soundscapes for a wide range of scenes, from natural landscapes to animated shorts. 🔹Balanced Multimodal Response: Our innovative multimodal diffusion transformer (MMDiT) architecture ensures the model balances video and text cues, generating rich, layered sound effects that capture every detail—from the main subject to subtle background elements. 🔹High-Fidelity Audio: Using a Representation Alignment (REPA) loss function and a powerful Audio VAE, we've improved generation stability and producing professional-grade audio, free of noise and inconsistencies. HunyuanVideo-Foley achieves SOTA on multiple benchmarks, surpassing all open-source models in audio quality, visual-semantic alignment, and temporal alignment. 👉Try it now: 🌐Project Page: 🔗Code: 📄Technical Report: 🤗Hugging Face:

Tencent Hy

122,539 次观看 • 10 个月前

🚨 JUST IN: MICROSOFT just open sourced a VOICE AI THAT TRANSCRIBES 60 MINUTES OF AUDIO in a single pass. 100% FREE. It knows who spoke. It knows when they spoke. It knows exactly what they said. All in one shot. No chunking. No context loss. It's called VibeVoice. Not a transcription tool. Not a basic speech to text wrapper. A frontier voice AI family with ASR, TTS, and real time streaming. All open source. All free. Here's what it actually does 👇 VibeVoice ASR - Speech Recognition: → Processes 60 minutes of continuous audio in a single pass → Never slices audio into chunks so global context is never lost → Identifies WHO spoke, WHEN they spoke and WHAT they said simultaneously → Supports customized hotwords for domain specific accuracy → Works in 50+ languages natively → Already adopted by Hugging Face Transformers library → Already being built on by the open source community BY PEOPLE WHO HAD NO IDEA THIS LEVEL OF ACCURACY WAS ALREADY FREE. VibeVoice TTS - Text to Speech: → Generates up to 90 minutes of speech in a single pass → Supports up to 4 distinct speakers in one conversation → Natural turn taking and speaker consistency throughout → Expressive speech that captures emotional nuances → Supports English, Chinese and multiple other languages VibeVoice Realtime - Streaming TTS: → Only 300 millisecond first audible latency → Streams text input in real time → 0.5B parameters so it actually deploys anywhere → Robust long form generation up to 10 minutes → Lightweight enough for production use today The core innovation nobody is talking about: Most voice AI models slice long audio into short chunks. Every time they slice, they lose context. Speaker tracking breaks. Semantic coherence breaks. Accuracy drops. VibeVoice uses continuous speech tokenizers running at an ultra low frame rate of 7.5 Hz. This preserves audio fidelity while dramatically boosting computational efficiency. The entire 60 minutes stays in context. Nothing gets lost. Nobody gets misidentified. The numbers: → VibeVoice ASR 7B - available now on Hugging Face → VibeVoice Realtime 0.5B - try it on Colab right now → 50+ supported languages → 11 distinct English voice styles → 9 multilingual speaker voices → Already integrated into Hugging Face Transformers → Finetuning code now available The wildest part? A voice powered input method called Vibing just built itself on top of VibeVoice ASR. Available on macOS and Windows right now. The open source community is already shipping products on top of this. 100% Open Source. Free to use. Free to fine tune. Free to build on. 🔖 Save this before your competitors find it first. 👇

🚨 JUST IN: MICROSOFT just open sourced a VOICE AI THAT TRANSCRIBES 60 MINUTES OF AUDIO in a single pass. 100% FREE. It knows who spoke. It knows when they spoke. It knows exactly what they said. All in one shot. No chunking. No context loss. It's called VibeVoice. Not a transcription tool. Not a basic speech to text wrapper. A frontier voice AI family with ASR, TTS, and real time streaming. All open source. All free. Here's what it actually does 👇 VibeVoice ASR - Speech Recognition: → Processes 60 minutes of continuous audio in a single pass → Never slices audio into chunks so global context is never lost → Identifies WHO spoke, WHEN they spoke and WHAT they said simultaneously → Supports customized hotwords for domain specific accuracy → Works in 50+ languages natively → Already adopted by Hugging Face Transformers library → Already being built on by the open source community BY PEOPLE WHO HAD NO IDEA THIS LEVEL OF ACCURACY WAS ALREADY FREE. VibeVoice TTS - Text to Speech: → Generates up to 90 minutes of speech in a single pass → Supports up to 4 distinct speakers in one conversation → Natural turn taking and speaker consistency throughout → Expressive speech that captures emotional nuances → Supports English, Chinese and multiple other languages VibeVoice Realtime - Streaming TTS: → Only 300 millisecond first audible latency → Streams text input in real time → 0.5B parameters so it actually deploys anywhere → Robust long form generation up to 10 minutes → Lightweight enough for production use today The core innovation nobody is talking about: Most voice AI models slice long audio into short chunks. Every time they slice, they lose context. Speaker tracking breaks. Semantic coherence breaks. Accuracy drops. VibeVoice uses continuous speech tokenizers running at an ultra low frame rate of 7.5 Hz. This preserves audio fidelity while dramatically boosting computational efficiency. The entire 60 minutes stays in context. Nothing gets lost. Nobody gets misidentified. The numbers: → VibeVoice ASR 7B - available now on Hugging Face → VibeVoice Realtime 0.5B - try it on Colab right now → 50+ supported languages → 11 distinct English voice styles → 9 multilingual speaker voices → Already integrated into Hugging Face Transformers → Finetuning code now available The wildest part? A voice powered input method called Vibing just built itself on top of VibeVoice ASR. Available on macOS and Windows right now. The open source community is already shipping products on top of this. 100% Open Source. Free to use. Free to fine tune. Free to build on. 🔖 Save this before your competitors find it first. 👇

Kanika

220,656 次观看 • 3 个月前

Here are 10 AI video editor GitHub repos worth bookmarking: 1. Shotcut Most actively maintained open source video editor in 2026. 14K stars. Cross-platform with AI-assisted features. Just shipped a new release April 30, 2026. 2. Kdenlive The closest open source alternative to Adobe Premiere Pro. Multi-track editing, proxy editing, VST audio, and customizable workspace. Best for professional workflows. 3. OpenShot The easiest entry point for beginners. Drag and drop, 400+ transitions, 3D titles, and AI-assisted trimming. 5,700 stars. 4. Blender Not just 3D. Blender's video sequence editor and compositing pipeline is used in professional film production. 18,300 stars. Unmatched for VFX. 5. Recordly Screen recorder with auto-zoom, cursor polish, webcam overlays, and styled frames built in. Built for demo videos and walkthroughs. 6. Wan2.1 Alibaba's open source text-to-video model. Cinema-grade 1080p generation. Apache 2.0. The gold standard for open source video generation in 2026. 7. HunyuanVideo Tencent's 13B parameter open source video model. 11.9K stars. Handles 720p and 1080p with high temporal coherence. 8. CogVideoX Apache 2.0 licensed. Loads natively via Hugging Face Diffusers. Strong prompt following and smooth frame transitions. Needs 16GB VRAM minimum. 12.5K stars. 9. Open-Sora Most starred open source video generation project at 24K stars. Full training pipeline for $200K. Production-level output quality. 10. Mochi 1 Focused entirely on motion quality. The most natural-looking physics of any open source video model. Water, fabric, and human gestures without AI jitter. Apache 2.0.

Here are 10 AI video editor GitHub repos worth bookmarking: 1. Shotcut Most actively maintained open source video editor in 2026. 14K stars. Cross-platform with AI-assisted features. Just shipped a new release April 30, 2026. 2. Kdenlive The closest open source alternative to Adobe Premiere Pro. Multi-track editing, proxy editing, VST audio, and customizable workspace. Best for professional workflows. 3. OpenShot The easiest entry point for beginners. Drag and drop, 400+ transitions, 3D titles, and AI-assisted trimming. 5,700 stars. 4. Blender Not just 3D. Blender's video sequence editor and compositing pipeline is used in professional film production. 18,300 stars. Unmatched for VFX. 5. Recordly Screen recorder with auto-zoom, cursor polish, webcam overlays, and styled frames built in. Built for demo videos and walkthroughs. 6. Wan2.1 Alibaba's open source text-to-video model. Cinema-grade 1080p generation. Apache 2.0. The gold standard for open source video generation in 2026. 7. HunyuanVideo Tencent's 13B parameter open source video model. 11.9K stars. Handles 720p and 1080p with high temporal coherence. 8. CogVideoX Apache 2.0 licensed. Loads natively via Hugging Face Diffusers. Strong prompt following and smooth frame transitions. Needs 16GB VRAM minimum. 12.5K stars. 9. Open-Sora Most starred open source video generation project at 24K stars. Full training pipeline for $200K. Production-level output quality. 10. Mochi 1 Focused entirely on motion quality. The most natural-looking physics of any open source video model. Water, fabric, and human gestures without AI jitter. Apache 2.0.

Kanika

17,309 次观看 • 1 个月前

VoxCPM 2 just dropped by OpenBMB Only 2B-param open-source TTS (Text-to-Speech) model built for production-grade multilingual voice work. Apache-2.0 license, Can run on only 8GB VRAM. • Eliminates the "robotic" feel of traditional TTS, delivering prosody and emotional depth suitable for high-stakes professional environments like filmmaking, gaming, animation, and audiobooks. • 30-language multilingual: no language tag needed, just type in a supported language and generate directly. • Voice design: create a brand-new voice from a text description alone, like age, tone, pace, or emotion. No reference audio required. Describe the desired voice characteristics (gender, age, tone, emotion, pace …) in Control Instruction, and VoxCPM2 will craft a unique voice from your description alone. • Controllable cloning: clone from a short clip, then steer delivery style without losing the speaker’s core voice. • Ultimate cloning: use reference audio + transcript for continuation-style cloning that keeps the tiny vocal details. • 48kHz output: takes 16kHz reference audio and produces studio-quality speech without an external upsampler. • Real-time ready: around 0.3 RTF on RTX 4090, even lower with Nano-VLLM. • Commercial use: Apache-2.0 licensed. Developer-Friendly Infrastructure: - Native Torch Inference: Direct support for PyTorch-based workflows. - Training Flexibility: Supports both full-parameter and LoRA fine-tuning for specific domain adaptation. - Production Readiness: Compatible with voxcpm-nanovllm for large-scale, high-concurrency deployment.

Rohan Paul

13,541 次观看 • 3 个月前

Text-to-speech is moving way too fast. Just a few days ago, I tweeted about PersonaPlex-7B, NVIDIA's new open source TTS ( And today, Qwen just open-sourced Qwen3-TTS 🤯 It’s a revolutionary text-to-speech model built for control. Not just about generating speech, but about shaping how it sounds directly from language. You can guide the pace, the tone, and the expressiveness straight from text, without touching audio graphs or hand-tuning parameters. That’s the real shift! What makes Qwen3-TTS stand out is how practical it already is: → voice cloning from just a few seconds of audio → voice creation without any reference sample → support for 10 languages out of the box → end-to-end latency down to ~97ms → works in both streaming and non-streaming setups The models come in two sizes (0.6B and 1.7B), so you can trade off quality and hardware cost depending on your setup. You can work with curated voices, designed voices, or cloned ones, and it integrates cleanly with vLLM for production use. It also ships as a simple Python package you can pip install. If you’re building real-time voice systems, this removes a lot of friction! 100% free and open source. I put the repo in the 🧵↓

Text-to-speech is moving way too fast. Just a few days ago, I tweeted about PersonaPlex-7B, NVIDIA's new open source TTS ( And today, Qwen just open-sourced Qwen3-TTS 🤯 It’s a revolutionary text-to-speech model built for control. Not just about generating speech, but about shaping how it sounds directly from language. You can guide the pace, the tone, and the expressiveness straight from text, without touching audio graphs or hand-tuning parameters. That’s the real shift! What makes Qwen3-TTS stand out is how practical it already is: → voice cloning from just a few seconds of audio → voice creation without any reference sample → support for 10 languages out of the box → end-to-end latency down to ~97ms → works in both streaming and non-streaming setups The models come in two sizes (0.6B and 1.7B), so you can trade off quality and hardware cost depending on your setup. You can work with curated voices, designed voices, or cloned ones, and it integrates cleanly with vLLM for production use. It also ships as a simple Python package you can pip install. If you’re building real-time voice systems, this removes a lot of friction! 100% free and open source. I put the repo in the 🧵↓

Charly Wargnier

59,144 次观看 • 5 个月前

🚀 VoxCPM 2 is live! 🎉 Another open-source AI #TTS model from China — and one that stands shoulder to shoulder with Qwen3-TTS, while bringing everything into a single unified model. After rapid iterations from V1 (zero-shot cloning) to V1.5 (long-form + fine-tuning), #VoxCPM has consistently pushed quality and usability forward. Now, VoxCPM 2 takes it further: 🔹30+ languages — truly global, truly local. 🔹Infinite voice design — type it, hear it, control it. From a whisper to a booming cinematic voice. 🔹Studio-grade audio — 48kHz ultra-high fidelity with emotional depth 🔹Diffusion-Autoregressive cloning — preserves more acoustic and emotional detail than token-based models like Qwen3-TTS 💡 Big shoutout to Grok — used your multi-image video magic for our launch demo. It’s scarily good at keeping visuals consistent across shots. Elon Elon Musk, this one’s for you. 😉 Check the demo & start cloning your dream voice: 🌐 Hugging Face Space: 🤗 Hugging Face Model: 🤖 ModelScope Model: 💻 GitHub： #TTS #AI #VoiceCloning #GrokImagine #ElonMusk #OpenBMB #VoxCPM

🚀 VoxCPM 2 is live! 🎉 Another open-source AI #TTS model from China — and one that stands shoulder to shoulder with Qwen3-TTS, while bringing everything into a single unified model. After rapid iterations from V1 (zero-shot cloning) to V1.5 (long-form + fine-tuning), #VoxCPM has consistently pushed quality and usability forward. Now, VoxCPM 2 takes it further: 🔹30+ languages — truly global, truly local. 🔹Infinite voice design — type it, hear it, control it. From a whisper to a booming cinematic voice. 🔹Studio-grade audio — 48kHz ultra-high fidelity with emotional depth 🔹Diffusion-Autoregressive cloning — preserves more acoustic and emotional detail than token-based models like Qwen3-TTS 💡 Big shoutout to Grok — used your multi-image video magic for our launch demo. It’s scarily good at keeping visuals consistent across shots. Elon Elon Musk, this one’s for you. 😉 Check the demo & start cloning your dream voice: 🌐 Hugging Face Space: 🤗 Hugging Face Model: 🤖 ModelScope Model: 💻 GitHub： #TTS #AI #VoiceCloning #GrokImagine #ElonMusk #OpenBMB #VoxCPM

OpenBMB

557,835 次观看 • 3 个月前

Today, we’re launching Gemma 4, our most intelligent open models to date. Built with the same breakthrough technology as Gemini 3, Gemma 4 brings advanced reasoning to your personal hardware and devices. Here’s what Gemma 4 unlocks for developers: — Intelligence-per-parameter: Our 31B (Dense) and 26B (MoE) models deliver state-of-the-art performance for their size, outcompeting models 20x their size on Arena.ai — Commercial flexibility: Released under a permissive Apache 2.0 license for complete developer flexibility and digital sovereignty — Agentic workflows: Native support for function-calling and structured JSON output allows you to build reliable, autonomous agents — Multimodal edge AI: The E2B and E4B models bring native vision, audio, and low latency to mobile and IoT devices — Long-context reasoning: Up to 256K context windows allow you to process entire repositories or large documents in a single prompt Whether you're building global applications in 140+ languages or local-first AI code assistants, Gemma 4 is built to be your foundation. Explore in Google AI Studio or download the weights on Hugging Face, Kaggle, and ollama.

Today, we’re launching Gemma 4, our most intelligent open models to date. Built with the same breakthrough technology as Gemini 3, Gemma 4 brings advanced reasoning to your personal hardware and devices. Here’s what Gemma 4 unlocks for developers: — Intelligence-per-parameter: Our 31B (Dense) and 26B (MoE) models deliver state-of-the-art performance for their size, outcompeting models 20x their size on Arena.ai — Commercial flexibility: Released under a permissive Apache 2.0 license for complete developer flexibility and digital sovereignty — Agentic workflows: Native support for function-calling and structured JSON output allows you to build reliable, autonomous agents — Multimodal edge AI: The E2B and E4B models bring native vision, audio, and low latency to mobile and IoT devices — Long-context reasoning: Up to 256K context windows allow you to process entire repositories or large documents in a single prompt Whether you're building global applications in 140+ languages or local-first AI code assistants, Gemma 4 is built to be your foundation. Explore in Google AI Studio or download the weights on Hugging Face, Kaggle, and ollama.

Google AI

1,669,084 次观看 • 3 个月前

Alibaba just released a coding model that hits 82 percent on SWE-Bench Verified. That is the highest score ever published for an open-source model. The weights are free. The license is Apache 2.0. You can run it today. The model is Qwen 4 Coder 32B. Here is what 82 percent on SWE-Bench Verified actually means. SWE-Bench Verified tests whether an AI can autonomously resolve real bugs pulled from real production GitHub repositories. Not synthetic exercises. Real open-source projects that real teams depend on. A model gets a bug report, reads the code, writes a fix, and either passes the test suite or it does not. At 82 percent, Qwen 4 Coder 32B resolves 82 out of every 100 real production bugs it is given. Without a human guiding it. On code it has never seen before. For comparison: Qwen 4 Coder 32B: 82 percent SWE-Bench Verified. Open source. Apache 2.0. Claude Fable 5: 80.3 percent SWE-Bench Pro. $10 input / $50 output per million tokens. Currently suspended. GPT-5.6 Sol: Competitive on Terminal-Bench. $5 input / $30 output per million tokens. An open-weight model that you can download and run for free just beat both of them on the benchmark designed to measure real software engineering capability. Here is the architecture. Qwen 4 Coder 32B is a 32 billion parameter dense model. Not a Mixture-of-Experts. Every parameter is active on every request. This matters for inference: a dense 32B model runs on 22 gigabytes of VRAM, which fits on a single high-end consumer GPU or a MacBook Pro with 64GB of unified memory. The smaller variant, Qwen 4 Coder 4B, runs at approximately 135 tokens per second on an M5 Max and fits inside 8 gigabytes of RAM. For a model with usable coding capability, that is a new bar for what fits in a single laptop. The training methodology continued Alibaba's approach of reinforcement learning on verifiable coding tasks. The model gets rewarded when its code passes tests. It gets penalized when it fails. Over millions of training steps, the model learns to write code that actually runs rather than code that looks plausible. License: Apache 2.0. Full commercial use. No attribution requirement. No revenue threshold. No monthly active user ceiling. Weights: Hugging Face, available today. Runs on: vLLM, Ollama, SGLang, and any standard GGUF-compatible inference engine. Qwen 4 32B also runs at approximately 135 tokens per second on an M5 Max chip, setting a new bar for what a sub-8GB model can do on Apple Silicon. The open-source coding model just beat the best closed-source model in the world on the benchmark designed to test whether AI can actually do software engineering. The weights are free. The subscription is optional. Source: Autom8Labs AI Insight July 2026, State of Open Source LLMs June 2026, Kunal Ganglani blog June 2026.

Alibaba just released a coding model that hits 82 percent on SWE-Bench Verified. That is the highest score ever published for an open-source model. The weights are free. The license is Apache 2.0. You can run it today. The model is Qwen 4 Coder 32B. Here is what 82 percent on SWE-Bench Verified actually means. SWE-Bench Verified tests whether an AI can autonomously resolve real bugs pulled from real production GitHub repositories. Not synthetic exercises. Real open-source projects that real teams depend on. A model gets a bug report, reads the code, writes a fix, and either passes the test suite or it does not. At 82 percent, Qwen 4 Coder 32B resolves 82 out of every 100 real production bugs it is given. Without a human guiding it. On code it has never seen before. For comparison: Qwen 4 Coder 32B: 82 percent SWE-Bench Verified. Open source. Apache 2.0. Claude Fable 5: 80.3 percent SWE-Bench Pro. $10 input / $50 output per million tokens. Currently suspended. GPT-5.6 Sol: Competitive on Terminal-Bench. $5 input / $30 output per million tokens. An open-weight model that you can download and run for free just beat both of them on the benchmark designed to measure real software engineering capability. Here is the architecture. Qwen 4 Coder 32B is a 32 billion parameter dense model. Not a Mixture-of-Experts. Every parameter is active on every request. This matters for inference: a dense 32B model runs on 22 gigabytes of VRAM, which fits on a single high-end consumer GPU or a MacBook Pro with 64GB of unified memory. The smaller variant, Qwen 4 Coder 4B, runs at approximately 135 tokens per second on an M5 Max and fits inside 8 gigabytes of RAM. For a model with usable coding capability, that is a new bar for what fits in a single laptop. The training methodology continued Alibaba's approach of reinforcement learning on verifiable coding tasks. The model gets rewarded when its code passes tests. It gets penalized when it fails. Over millions of training steps, the model learns to write code that actually runs rather than code that looks plausible. License: Apache 2.0. Full commercial use. No attribution requirement. No revenue threshold. No monthly active user ceiling. Weights: Hugging Face, available today. Runs on: vLLM, Ollama, SGLang, and any standard GGUF-compatible inference engine. Qwen 4 32B also runs at approximately 135 tokens per second on an M5 Max chip, setting a new bar for what a sub-8GB model can do on Apple Silicon. The open-source coding model just beat the best closed-source model in the world on the benchmark designed to test whether AI can actually do software engineering. The weights are free. The subscription is optional. Source: Autom8Labs AI Insight July 2026, State of Open Source LLMs June 2026, Kunal Ganglani blog June 2026.

Harman

38,953 次观看 • 7 天前

🚀 🚀Excited to announce the technical report of MiniCPM-o 4.5! MiniCPM-o 4.5 transitions #AI interaction from traditional turn-based processing to a real-time, native full-duplex stream-based paradigm. 🌊 The Omni-Flow Framework Instead of traditional VAD-based workarounds, we introduce the #Omni-#Flow framework. This unified stream paradigm aligns video, audio, and text on a synchronized millisecond timeline. • Native Full-Duplex: Simultaneous perception and response. • Proactive Interaction: Natively manages turn-taking without external VAD, supports proactive reminding. 📉 9B Scale, SOTA Performance MiniCPM-o 4.5 demonstrates SOTA multimodal intelligence at its scale: • Multimodal Benchmarks: Comparable to #Gemini 2.5 Flash on MMBench EN (87.6) and MathVista (80.1). • Streaming Evaluation: 54.4% win rate on LiveSports-3K-CC, surpassing specialized models. 💻 The Ultimate Edge AI — Fully Functional without Network Connection We are providing one-click installers for Windows (12G VRAM,RTX 5070) and macOS (M1-M5 Max/ M5 Pro). • Local API Support: Deploy your own inference server to integrate native full-duplex into custom apps. • Free Access: We are offering free community API services for exploration. • 100% Private: Your data never leaves your machine. Deploy in under 10 minutes. 🛠️👇 👐 Join the Open Future The weights are open. The protocol is public. 📄 Technical Report: 💻 GitHub: 🤗 HuggingFace: 🌐 Web Demo: #MiniCPMo #OpenSourceAI #EdgeAI #MachineLearning #ComputerVision #LLM

🚀 🚀Excited to announce the technical report of MiniCPM-o 4.5! MiniCPM-o 4.5 transitions #AI interaction from traditional turn-based processing to a real-time, native full-duplex stream-based paradigm. 🌊 The Omni-Flow Framework Instead of traditional VAD-based workarounds, we introduce the #Omni-#Flow framework. This unified stream paradigm aligns video, audio, and text on a synchronized millisecond timeline. • Native Full-Duplex: Simultaneous perception and response. • Proactive Interaction: Natively manages turn-taking without external VAD, supports proactive reminding. 📉 9B Scale, SOTA Performance MiniCPM-o 4.5 demonstrates SOTA multimodal intelligence at its scale: • Multimodal Benchmarks: Comparable to #Gemini 2.5 Flash on MMBench EN (87.6) and MathVista (80.1). • Streaming Evaluation: 54.4% win rate on LiveSports-3K-CC, surpassing specialized models. 💻 The Ultimate Edge AI — Fully Functional without Network Connection We are providing one-click installers for Windows (12G VRAM,RTX 5070) and macOS (M1-M5 Max/ M5 Pro). • Local API Support: Deploy your own inference server to integrate native full-duplex into custom apps. • Free Access: We are offering free community API services for exploration. • 100% Private: Your data never leaves your machine. Deploy in under 10 minutes. 🛠️👇 👐 Join the Open Future The weights are open. The protocol is public. 📄 Technical Report: 💻 GitHub: 🤗 HuggingFace: 🌐 Web Demo: #MiniCPMo #OpenSourceAI #EdgeAI #MachineLearning #ComputerVision #LLM

OpenBMB

147,824 次观看 • 2 个月前

six months ago this wasn't happening on 8gb vram. running unsloth's Q4_K_XL quant of gemma 4 26b-a4b-it-qat, a sparse MoE model with only 4b active params on a single rtx 4060 laptop gpu, 8gb vram, 20+ tok/s decode. no cloud, no api, no offload hacks. just a gaming laptop on battery. what makes it fit: google's QAT (quantization aware training), plus MTP (multi token prediction) support in the latest llama.cpp builds. that combo is the single biggest unlock for local inference on low vram. rtx 3060, rtx 3070, gtx 1070, gtx 1080, rtx 4050, rtx 4060, rtx 5050, rtx 5060 — any 6-8gb consumer gpu, old or new — this model runs on it. world cup season, so i told it to build a soccer themed flappy bird clone. one shot, zero iteration, fully playable. six months ago an 8gb model could barely clone vanilla flappy bird. now it's shipping a themed game from a sparse MoE model running locally on a laptop battery. inference benchmarks: - decode throughput: 30 tok/s - context: 64k. this is the real unlock. 64k ctx is what makes a hermes agent loop viable locally on this model, not just single-turn chat. llama.cpp flags: -m gemma-4-26B-A4B-it-qat-UD-Q4_K_XL.gguf -c 64000 -cmoe --port 8080 game's deployed on my own site, built and shipped end to end with open source llm, zero closed source api dependency in the pipeline. link in the description. gguf weights on huggingface, link in the comments. pull it down, run it on whatever 8gb card is sitting in your rig. try the game and tell me your score and what you want in v2. local llms on consumer gpus stopped being a meme.

six months ago this wasn't happening on 8gb vram. running unsloth's Q4_K_XL quant of gemma 4 26b-a4b-it-qat, a sparse MoE model with only 4b active params on a single rtx 4060 laptop gpu, 8gb vram, 20+ tok/s decode. no cloud, no api, no offload hacks. just a gaming laptop on battery. what makes it fit: google's QAT (quantization aware training), plus MTP (multi token prediction) support in the latest llama.cpp builds. that combo is the single biggest unlock for local inference on low vram. rtx 3060, rtx 3070, gtx 1070, gtx 1080, rtx 4050, rtx 4060, rtx 5050, rtx 5060 — any 6-8gb consumer gpu, old or new — this model runs on it. world cup season, so i told it to build a soccer themed flappy bird clone. one shot, zero iteration, fully playable. six months ago an 8gb model could barely clone vanilla flappy bird. now it's shipping a themed game from a sparse MoE model running locally on a laptop battery. inference benchmarks: - decode throughput: 30 tok/s - context: 64k. this is the real unlock. 64k ctx is what makes a hermes agent loop viable locally on this model, not just single-turn chat. llama.cpp flags: -m gemma-4-26B-A4B-it-qat-UD-Q4_K_XL.gguf -c 64000 -cmoe --port 8080 game's deployed on my own site, built and shipped end to end with open source llm, zero closed source api dependency in the pipeline. link in the description. gguf weights on huggingface, link in the comments. pull it down, run it on whatever 8gb card is sitting in your rig. try the game and tell me your score and what you want in v2. local llms on consumer gpus stopped being a meme.

Alok

60,725 次观看 • 23 天前

We've officially released and open-sourced HunyuanImage 2.1, our latest text-to-image model. The new model delivers on our commitment to balancing performance and quality. With native 2K image generation, HunyuanImage 2.1 is an advanced open-source text-to-image model.🎨 ✨ New in 2.1: 🔹Advanced Semantics: Supports ultra-long and complex prompts of up to 1000 tokens, and precisely controls the generation of multiple subjects in a single image. 🔹Precise Chinese and English Text Rendering with seamless image–text integration: The model naturally integrates text into images, making it suitable for a wide range of applications such as product covers, illustrations, and poster design to meet the needs of various fields. 🔹Rich Styles and High Aesthetic: Capable of generating images in various styles—including photorealistic portraits, comics, and vinyl figures—it delivers outstanding visual appeal and artistic quality. 🔹High-Quality Generation: Efficiently produces ultra-high-definition (2K) images in the same time other models take to generate a 1K image. HunyuanImage 2.1 uses two text encoders: a multimodal large language model (MLLM) to improve the model's image and text alignment capabilities, and a multi-language character-aware encoder to improve text rendering capabilities. The model is a single- and double-stream diffusion transformer with 17B parameters. We've also open-sourced the weights of the the accelerated version with meanflow which reduces inference steps from 100 to just 8, and PromptEnhancer, the first industrial-grade rewriting model that enhances your prompts for more nuanced and expressive image generation. Now, creators turn complex ideas—like posters with slogans or multi-panel comics—into visuals faster than ever. We’re just getting started. Stay tuned for our native multimodal image generation model coming soon. 🌐Website: 🔗Github: 🤗Hugging Face: ✨Hugging Face Demo:

We've officially released and open-sourced HunyuanImage 2.1, our latest text-to-image model. The new model delivers on our commitment to balancing performance and quality. With native 2K image generation, HunyuanImage 2.1 is an advanced open-source text-to-image model.🎨 ✨ New in 2.1: 🔹Advanced Semantics: Supports ultra-long and complex prompts of up to 1000 tokens, and precisely controls the generation of multiple subjects in a single image. 🔹Precise Chinese and English Text Rendering with seamless image–text integration: The model naturally integrates text into images, making it suitable for a wide range of applications such as product covers, illustrations, and poster design to meet the needs of various fields. 🔹Rich Styles and High Aesthetic: Capable of generating images in various styles—including photorealistic portraits, comics, and vinyl figures—it delivers outstanding visual appeal and artistic quality. 🔹High-Quality Generation: Efficiently produces ultra-high-definition (2K) images in the same time other models take to generate a 1K image. HunyuanImage 2.1 uses two text encoders: a multimodal large language model (MLLM) to improve the model's image and text alignment capabilities, and a multi-language character-aware encoder to improve text rendering capabilities. The model is a single- and double-stream diffusion transformer with 17B parameters. We've also open-sourced the weights of the the accelerated version with meanflow which reduces inference steps from 100 to just 8, and PromptEnhancer, the first industrial-grade rewriting model that enhances your prompts for more nuanced and expressive image generation. Now, creators turn complex ideas—like posters with slogans or multi-panel comics—into visuals faster than ever. We’re just getting started. Stay tuned for our native multimodal image generation model coming soon. 🌐Website: 🔗Github: 🤗Hugging Face: ✨Hugging Face Demo:

Tencent Hy

89,257 次观看 • 10 个月前

We are excited to share #PDF2Audio, an open-source alternative to the #podcast feature of #NotebookLM with flexibility & tailored outputs that you can precisely control in the app: You can make a podcast, lecture, discussions, short/long form summaries & more, including the use of the amazing🍓o1 model (Sam Altman OpenAI: with stunning results!). Code & HF Space: You can find #PDF2Audio on GitHub for local use or try the Hugging Face space, all featuring Gradio. Link to the repo & HF space in the reply. Thank you @knowsuchagencyfor the great work on #promptic and #pdf2podcast, as well as LiteLLM (YC W23), & AK for helping us with the Hugging Face spaces version. We hope that this tool is useful for the community. Background: Developing audio podcasts, lectures, & summaries from complex documents & data has become an exciting trend with impacts from research to education to business. Our open-source #PDF2Audio tool that allows users to utilize various models such as #o1 or local/open-source models, to develop deep-dives into technical content. Example application - material design analysis: As an example to show what the system can do, check the video for a detailed 13-minute analysis of one of the designs created by #SciAgents merging silk & dandelion pigments, created using 🍓o1. The conversation describes the new material, an integration of silk proteins & luteolin/dandelion pigments to create a new biomaterial. Silk, a natural #nanostructured protein-based fiber known for its strength & flexibility, is combined with dandelion pigments like luteolin, which offer unique optical properties. By merging these components at the nanoscale level, the resulting material displays structural coloration—vibrant, tunable colors created by the material's structure rather than synthetic dyes, and leverages silk's hierarchical organization as a scaffold for the pigments, ensuring uniform distribution and non-covalent bonding at the molecular level. Key technical features include: ➡️Low-temperature processing to maintain the integrity of both silk and pigments while reducing energy consumption by 30%. ➡️Enhanced mechanical properties, with tensile strength up to 1.5 gigapascals. ➡️Potential self-healing capabilities and environmental responsiveness, allowing the material to repair minor damage and change color based on environmental conditions. ➡️UV protection and antimicrobial properties, which make this material ideal for smart textiles, eco-friendly coatings, and medical applications. This development opens new doors for sustainable materials, offering an eco-friendly alternative to synthetic fibers with applications in various industries, from fashion to healthcare.

We are excited to share #PDF2Audio, an open-source alternative to the #podcast feature of #NotebookLM with flexibility & tailored outputs that you can precisely control in the app: You can make a podcast, lecture, discussions, short/long form summaries & more, including the use of the amazing🍓o1 model (Sam Altman OpenAI: with stunning results!). Code & HF Space: You can find #PDF2Audio on GitHub for local use or try the Hugging Face space, all featuring Gradio. Link to the repo & HF space in the reply. Thank you @knowsuchagencyfor the great work on #promptic and #pdf2podcast, as well as LiteLLM (YC W23), & AK for helping us with the Hugging Face spaces version. We hope that this tool is useful for the community. Background: Developing audio podcasts, lectures, & summaries from complex documents & data has become an exciting trend with impacts from research to education to business. Our open-source #PDF2Audio tool that allows users to utilize various models such as #o1 or local/open-source models, to develop deep-dives into technical content. Example application - material design analysis: As an example to show what the system can do, check the video for a detailed 13-minute analysis of one of the designs created by #SciAgents merging silk & dandelion pigments, created using 🍓o1. The conversation describes the new material, an integration of silk proteins & luteolin/dandelion pigments to create a new biomaterial. Silk, a natural #nanostructured protein-based fiber known for its strength & flexibility, is combined with dandelion pigments like luteolin, which offer unique optical properties. By merging these components at the nanoscale level, the resulting material displays structural coloration—vibrant, tunable colors created by the material's structure rather than synthetic dyes, and leverages silk's hierarchical organization as a scaffold for the pigments, ensuring uniform distribution and non-covalent bonding at the molecular level. Key technical features include: ➡️Low-temperature processing to maintain the integrity of both silk and pigments while reducing energy consumption by 30%. ➡️Enhanced mechanical properties, with tensile strength up to 1.5 gigapascals. ➡️Potential self-healing capabilities and environmental responsiveness, allowing the material to repair minor damage and change color based on environmental conditions. ➡️UV protection and antimicrobial properties, which make this material ideal for smart textiles, eco-friendly coatings, and medical applications. This development opens new doors for sustainable materials, offering an eco-friendly alternative to synthetic fibers with applications in various industries, from fashion to healthcare.

Markus J. Buehler

208,353 次观看 • 1 年前

I've been building a music player with Next.js for fun. Here's a quick demo of how it works (it's open source!) • Demo: • Code: If you want to learn more about how it's built, here's more details ↓ I'm using Postgres (with Drizzle) to store information about the songs and playlists. Audio and image files are stored in Vercel Blob (object storage), and the URLs are then referenced in the database. For the UI, I'm using shadcn/ui (so Tailwind CSS and Radix). This made it easy to copy/paste in some nice components, like the dropdown menus. I built the entire first version of the UI in v0 and then iterated from there, feeding it my Drizzle schema as a source in the project and having it scaffold some of the boilerplate for me: I added support for keyboard navigation (using arrow keys) or vim motions (j/k to go up/down, and h/l to go between playlists and tracks). Also, space to toggle the now playing song, and / to focus the search input. The search function has a nice utility to highlight the currently searched text on the page in yellow. Then, I was exploring how to pass metadata from my application to macOS or iOS. Turns out there's an API for that – MediaSession. Web apps can share metadata about what media is playing (title, artist, album artwork) and sync play/pause/seek with system media controls. Works across modern browsers — even integrates with iOS dynamic island and shows up on lock screens: I set up my app like a PWA – it has a manifest.json file, so it can be installed to my iOS home screen or added to my dock on macOS. On iOS, it then uses the full screen height `100dvh` (dynamic viewport) and has padding on the bottom for the safe area with the `env()` CSS function. Finally, I was able to use the Vercel AI SDK in a script to clean up the metadata on audio files I downloaded from YouTube. Bonus: I even was able to dogfood the React Compiler, which helped me fix a performance bug! That's all! It's fun to make personal software:

I've been building a music player with Next.js for fun. Here's a quick demo of how it works (it's open source!) • Demo: • Code: If you want to learn more about how it's built, here's more details ↓ I'm using Postgres (with Drizzle) to store information about the songs and playlists. Audio and image files are stored in Vercel Blob (object storage), and the URLs are then referenced in the database. For the UI, I'm using shadcn/ui (so Tailwind CSS and Radix). This made it easy to copy/paste in some nice components, like the dropdown menus. I built the entire first version of the UI in v0 and then iterated from there, feeding it my Drizzle schema as a source in the project and having it scaffold some of the boilerplate for me: I added support for keyboard navigation (using arrow keys) or vim motions (j/k to go up/down, and h/l to go between playlists and tracks). Also, space to toggle the now playing song, and / to focus the search input. The search function has a nice utility to highlight the currently searched text on the page in yellow. Then, I was exploring how to pass metadata from my application to macOS or iOS. Turns out there's an API for that – MediaSession. Web apps can share metadata about what media is playing (title, artist, album artwork) and sync play/pause/seek with system media controls. Works across modern browsers — even integrates with iOS dynamic island and shows up on lock screens: I set up my app like a PWA – it has a manifest.json file, so it can be installed to my iOS home screen or added to my dock on macOS. On iOS, it then uses the full screen height `100dvh` (dynamic viewport) and has padding on the bottom for the safe area with the `env()` CSS function. Finally, I was able to use the Vercel AI SDK in a script to clean up the metadata on audio files I downloaded from YouTube. Bonus: I even was able to dogfood the React Compiler, which helped me fix a performance bug! That's all! It's fun to make personal software:

Lee Robinson

117,624 次观看 • 1 年前

no money for grok or midjourney? this tool is for you. there's a FREE tool created by an anon dev. open-source. runs locally. 117k stars on github. it generates: > images & video > 3d models > audio > 20+ models here's how to set it up in under 5 minutes: 1️⃣download ComfyUI Desktop go to and grab the desktop app for your system. windows 10+, mac (apple silicon), or linux. it installs like any normal app, it sets up python and every dependency for you in the background. no terminal, no config files. 2️⃣open it first launch, it spins up its own environment automatically. you just wait a few seconds and you're in. you'll land on a node canvas, that's the whole interface. 3️⃣load a starter workflow top menu → Workflow → Browse Templates → Image Generation. click it. this drops a ready-made setup onto your canvas so you don't build anything from scratch. 4️⃣grab a model comfyui ships empty on purpose, the model is the brain, and you pick it. in the template, the "Load Checkpoint" node has a Download button when no model is installed. click it. it pulls one in for you (a few GB, this is the only real wait). 5️⃣install ComfyUI Manager this is the one add-on you don't skip. it lets you install models, custom nodes, and updates with a click instead of the command line. grab it from github (link in comments). it's the difference between fighting comfyui and flying in it. one honest note: an NVIDIA gpu makes this fast, apple silicon works great too, and a weak machine still runs it just slower. that's the whole setup. you now own an image, video, and 3D studio that costs you nothing per month. save this. and the next time grok or midjourney asks for your card. you won't need it. disclaimer: comfyui itself is 100% free. so are the local models (sdxl, flux, wan 2.2, ltx-2). some premium models like seedance are pay-per-use api models, only if you want top-tier quality. the free local ones cover most of what you need. (github link in the comments) follow and turn on post notification for daily AI contents.

no money for grok or midjourney? this tool is for you. there's a FREE tool created by an anon dev. open-source. runs locally. 117k stars on github. it generates: > images & video > 3d models > audio > 20+ models here's how to set it up in under 5 minutes: 1️⃣download ComfyUI Desktop go to and grab the desktop app for your system. windows 10+, mac (apple silicon), or linux. it installs like any normal app, it sets up python and every dependency for you in the background. no terminal, no config files. 2️⃣open it first launch, it spins up its own environment automatically. you just wait a few seconds and you're in. you'll land on a node canvas, that's the whole interface. 3️⃣load a starter workflow top menu → Workflow → Browse Templates → Image Generation. click it. this drops a ready-made setup onto your canvas so you don't build anything from scratch. 4️⃣grab a model comfyui ships empty on purpose, the model is the brain, and you pick it. in the template, the "Load Checkpoint" node has a Download button when no model is installed. click it. it pulls one in for you (a few GB, this is the only real wait). 5️⃣install ComfyUI Manager this is the one add-on you don't skip. it lets you install models, custom nodes, and updates with a click instead of the command line. grab it from github (link in comments). it's the difference between fighting comfyui and flying in it. one honest note: an NVIDIA gpu makes this fast, apple silicon works great too, and a weak machine still runs it just slower. that's the whole setup. you now own an image, video, and 3D studio that costs you nothing per month. save this. and the next time grok or midjourney asks for your card. you won't need it. disclaimer: comfyui itself is 100% free. so are the local models (sdxl, flux, wan 2.2, ltx-2). some premium models like seedance are pay-per-use api models, only if you want top-tier quality. the free local ones cover most of what you need. (github link in the comments) follow and turn on post notification for daily AI contents.

m0h

14,542 次观看 • 1 个月前

10 free Google AI tools nobody talks about. while everyone's burning $20/mo on chatgpt and claude, google quietly shipped a stack worth $200+/mo. all free. all yours. — 1️⃣ NotebookLM — your second brain upload sources (PDFs, websites, audio, YouTube). it summarizes, builds mind maps, generates quizzes, drafts slide decks, even turns your notes into a podcast you can listen to on a walk. free tier: 100 notebooks, 50 sources each, 50 chats/day, 3 audio overviews/day. replaces: notion AI + perplexity + readwise — 2️⃣ Google AI Studio — the free gemini playground web playground for gemini 3 pro and flash with a free API key. generous limits. paste a 1M-token context window and watch it actually use it. faster than the openai playground and free where openai charges per token. replaces: openai playground + paid API credits — 3️⃣ Gemini CLI — google's open-source terminal agent apache 2.0 licensed. one command (npx @google/gemini-cli) and you've got an agent in your terminal that reads your codebase, runs shell commands, and ships PRs. drop-in claude code alternative. replaces: claude code ($20/mo by default) — 4️⃣ Jules — async coding agent assign jules a github issue. it spins up a cloud VM, clones your repo, writes the plan, makes the changes, opens a PR. free tier: 15 tasks/day, 3 concurrent, runs on gemini flash. replaces: devin ($20/mo+) + cursor agent 5️⃣ Stitch — text → UI → code google's free figma killer. describe an interface, get production-ready HTML/CSS/Tailwind + figma export. march 2026 update added voice canvas, infinite canvas, and MCP integration with cursor. 350 standard + 200 experimental generations/month free. replaces: galileo AI + early-stage figma work — 6️⃣ Gemma 4 — open-weight LLM google's flagship open model. apache 2.0. 2B, 4B, 26B-MoE, and 31B variants. 256K context. runs on ollama with one command. quantized versions run on a 4090 or beefy laptop. replaces: paying for hosted LLM inference — 7️⃣ Illuminate — papers → podcasts paste an arxiv preprint link. illuminate turns dense research papers into a 6-8 min conversation between two AI hosts breaking it down. perfect for commute reading you can't do at a desk. note: still in waitlist for some regions. replaces: snipd + manual research reading — 8️⃣ Learn About (LearnLM) — adaptive AI tutor drop in any topic you're stuck on. highlight a word, click "go deeper," and the interface adapts in real time to your comprehension level. visual explanations, follow-up questions, the works. replaces: paid tutoring on niche topics — 9️⃣ Google Labs FX (ImageFX + Flow + MusicFX) — free imagen, veo, musicLM google labs creative suite. text-to-image (imagen 4), text-to-video (veo via Flow), text-to-music (musicLM). free tier: limited daily generations. the heavy veo 3.1 features are paid (AI Pro $19.99/mo). still worth using for image and music — those stay free. replaces: midjourney + suno (free tier only — runway-level video gen is paid) — 🔟 Google Colab — free GPU notebooks free T4 GPU + 12GB RAM in a browser tab. enough to fine-tune small models, run stable diffusion, prototype agents. the launching pad for half the ML projects on github. replaces: paid cloud GPU rentals — a quick honest note: these tools aren't 1:1 better than the paid versions they replace. but they're decent enough to get most things done — especially if you're not a heavy user or you've got little funds to play with. i've put all 10 in a public github repo (link in comments). follow + turn on post notifications for more useful posts like this 🔔

10 free Google AI tools nobody talks about. while everyone's burning $20/mo on chatgpt and claude, google quietly shipped a stack worth $200+/mo. all free. all yours. — 1️⃣ NotebookLM — your second brain upload sources (PDFs, websites, audio, YouTube). it summarizes, builds mind maps, generates quizzes, drafts slide decks, even turns your notes into a podcast you can listen to on a walk. free tier: 100 notebooks, 50 sources each, 50 chats/day, 3 audio overviews/day. replaces: notion AI + perplexity + readwise — 2️⃣ Google AI Studio — the free gemini playground web playground for gemini 3 pro and flash with a free API key. generous limits. paste a 1M-token context window and watch it actually use it. faster than the openai playground and free where openai charges per token. replaces: openai playground + paid API credits — 3️⃣ Gemini CLI — google's open-source terminal agent apache 2.0 licensed. one command (npx @google/gemini-cli) and you've got an agent in your terminal that reads your codebase, runs shell commands, and ships PRs. drop-in claude code alternative. replaces: claude code ($20/mo by default) — 4️⃣ Jules — async coding agent assign jules a github issue. it spins up a cloud VM, clones your repo, writes the plan, makes the changes, opens a PR. free tier: 15 tasks/day, 3 concurrent, runs on gemini flash. replaces: devin ($20/mo+) + cursor agent 5️⃣ Stitch — text → UI → code google's free figma killer. describe an interface, get production-ready HTML/CSS/Tailwind + figma export. march 2026 update added voice canvas, infinite canvas, and MCP integration with cursor. 350 standard + 200 experimental generations/month free. replaces: galileo AI + early-stage figma work — 6️⃣ Gemma 4 — open-weight LLM google's flagship open model. apache 2.0. 2B, 4B, 26B-MoE, and 31B variants. 256K context. runs on ollama with one command. quantized versions run on a 4090 or beefy laptop. replaces: paying for hosted LLM inference — 7️⃣ Illuminate — papers → podcasts paste an arxiv preprint link. illuminate turns dense research papers into a 6-8 min conversation between two AI hosts breaking it down. perfect for commute reading you can't do at a desk. note: still in waitlist for some regions. replaces: snipd + manual research reading — 8️⃣ Learn About (LearnLM) — adaptive AI tutor drop in any topic you're stuck on. highlight a word, click "go deeper," and the interface adapts in real time to your comprehension level. visual explanations, follow-up questions, the works. replaces: paid tutoring on niche topics — 9️⃣ Google Labs FX (ImageFX + Flow + MusicFX) — free imagen, veo, musicLM google labs creative suite. text-to-image (imagen 4), text-to-video (veo via Flow), text-to-music (musicLM). free tier: limited daily generations. the heavy veo 3.1 features are paid (AI Pro $19.99/mo). still worth using for image and music — those stay free. replaces: midjourney + suno (free tier only — runway-level video gen is paid) — 🔟 Google Colab — free GPU notebooks free T4 GPU + 12GB RAM in a browser tab. enough to fine-tune small models, run stable diffusion, prototype agents. the launching pad for half the ML projects on github. replaces: paid cloud GPU rentals — a quick honest note: these tools aren't 1:1 better than the paid versions they replace. but they're decent enough to get most things done — especially if you're not a heavy user or you've got little funds to play with. i've put all 10 in a public github repo (link in comments). follow + turn on post notifications for more useful posts like this 🔔

m0h

11,673 次观看 • 1 个月前

10 free github repos that can replace major SaaS with subscriptions. all free. open-sourced. some are MIT licensed. — 1️⃣ openscreen — replaces screen studio ($29/mo) - a clean macOS/windows/linux screen recorder for polished demos. - blur, cursor highlighting, annotations, export to mp4 or gif at any aspect ratio. - doesn't try to clone every feature, just nails the basics for quick walkthroughs you'd post on X. — 2️⃣ voicebox — replaces elevenlabs ($22/mo) + wisprflow ($15/mo) - local-first AI voice studio. - clone voices from 3 seconds of audio, generate speech across 7 TTS engines in 23 languages, - dictate into any text field with a global hotkey. - nothing leaves your machine. - runs on apple silicon, cuda, rocm. — 3️⃣ openshorts — replaces opus clip ($19/mo) + submagic ($16/mo) - free AI video platform. - clip generator turns long youtube videos into 9:16 shorts with auto-subtitles and face tracking (runs on free gemini + elevenlabs tiers). - also includes AI UGC video generation with actors — that part is pay-per-use via fal. ai (~$0.65-2 per video). docker self-host. — 4️⃣ freellmapi — replaces chatgpt pro + claude pro ($20/mo each) - stacks 14 free AI provider tiers (google, groq, cerebras, openrouter, github models + 9 more) behind one openai-compatible endpoint. ~800M tokens/month. - smart router with failover, sticky sessions, encrypted key storage. ships with a dashboard. — 5️⃣ playwright-mcp — replaces browserbase ($39/mo) + browser use ($25/mo) - microsoft's official MCP server that gives any AI agent full browser control. - uses accessibility trees, not screenshots — deterministic and token-efficient. - works with claude code, cursor, windsurf, codex out of the box. — 6️⃣ vibe-trading — replaces tradingview premium ($60/mo) - natural-language finance research agent. - 7 backtest engines across stocks, crypto, futures, forex. - 75 specialist skills (factor analysis, options strategy, ML strategy). - 29 multi-agent swarm presets. - 21 of 22 MCP tools work with zero API keys. — 7️⃣ CalCom — replaces calendly ($12/mo) + savvycal ($12/mo) - the open-source scheduling infrastructure. - one-on-ones, group events, round-robin, team booking, - payment collection (stripe), routing forms, workflows. - integrates with google/outlook/apple calendar, zoom, meet, teams. - self-host in 10 minutes with docker. 40k stars. — 8️⃣ whisper — replaces otter ($17/mo) - openAI's open-source speech-to-text model. - transcribe audio in 99 languages, translate to english, generate timestamps. - runs locally on cpu or gpu. - the actual model behind most "AI transcription" SaaS tools you're paying for. — 9️⃣ postiz — replaces buffer ($15/mo) - AI-powered social media scheduler. - cross-post to X, linkedin, instagram, tiktok, threads, bluesky, mastodon, youtube, pinterest. - AI captions and hashtags. - analytics dashboard. team workspaces. 31k stars and rising. — 🔟 vaultwarden — replaces 1password ($8/mo) - unofficial bitwarden-compatible server written in rust. - works with every official bitwarden client (mobile, desktop, browser). - unlimited users, unlimited vaults, full enterprise feature set. - runs on a $5 VPS or your home server. — disclaimer: open-source ≠ 1:1 replacement. you'll trade polish for ownership, hand-holding for control, and a credit card for a github version. for builders, prototypers, and indie hackers — that's the whole point. for everyone else, the paid tools still have their place. bookmark this. share with one friend bleeding subscription fees. ~m0h

10 free github repos that can replace major SaaS with subscriptions. all free. open-sourced. some are MIT licensed. — 1️⃣ openscreen — replaces screen studio ($29/mo) - a clean macOS/windows/linux screen recorder for polished demos. - blur, cursor highlighting, annotations, export to mp4 or gif at any aspect ratio. - doesn't try to clone every feature, just nails the basics for quick walkthroughs you'd post on X. — 2️⃣ voicebox — replaces elevenlabs ($22/mo) + wisprflow ($15/mo) - local-first AI voice studio. - clone voices from 3 seconds of audio, generate speech across 7 TTS engines in 23 languages, - dictate into any text field with a global hotkey. - nothing leaves your machine. - runs on apple silicon, cuda, rocm. — 3️⃣ openshorts — replaces opus clip ($19/mo) + submagic ($16/mo) - free AI video platform. - clip generator turns long youtube videos into 9:16 shorts with auto-subtitles and face tracking (runs on free gemini + elevenlabs tiers). - also includes AI UGC video generation with actors — that part is pay-per-use via fal. ai (~$0.65-2 per video). docker self-host. — 4️⃣ freellmapi — replaces chatgpt pro + claude pro ($20/mo each) - stacks 14 free AI provider tiers (google, groq, cerebras, openrouter, github models + 9 more) behind one openai-compatible endpoint. ~800M tokens/month. - smart router with failover, sticky sessions, encrypted key storage. ships with a dashboard. — 5️⃣ playwright-mcp — replaces browserbase ($39/mo) + browser use ($25/mo) - microsoft's official MCP server that gives any AI agent full browser control. - uses accessibility trees, not screenshots — deterministic and token-efficient. - works with claude code, cursor, windsurf, codex out of the box. — 6️⃣ vibe-trading — replaces tradingview premium ($60/mo) - natural-language finance research agent. - 7 backtest engines across stocks, crypto, futures, forex. - 75 specialist skills (factor analysis, options strategy, ML strategy). - 29 multi-agent swarm presets. - 21 of 22 MCP tools work with zero API keys. — 7️⃣ CalCom — replaces calendly ($12/mo) + savvycal ($12/mo) - the open-source scheduling infrastructure. - one-on-ones, group events, round-robin, team booking, - payment collection (stripe), routing forms, workflows. - integrates with google/outlook/apple calendar, zoom, meet, teams. - self-host in 10 minutes with docker. 40k stars. — 8️⃣ whisper — replaces otter ($17/mo) - openAI's open-source speech-to-text model. - transcribe audio in 99 languages, translate to english, generate timestamps. - runs locally on cpu or gpu. - the actual model behind most "AI transcription" SaaS tools you're paying for. — 9️⃣ postiz — replaces buffer ($15/mo) - AI-powered social media scheduler. - cross-post to X, linkedin, instagram, tiktok, threads, bluesky, mastodon, youtube, pinterest. - AI captions and hashtags. - analytics dashboard. team workspaces. 31k stars and rising. — 🔟 vaultwarden — replaces 1password ($8/mo) - unofficial bitwarden-compatible server written in rust. - works with every official bitwarden client (mobile, desktop, browser). - unlimited users, unlimited vaults, full enterprise feature set. - runs on a $5 VPS or your home server. — disclaimer: open-source ≠ 1:1 replacement. you'll trade polish for ownership, hand-holding for control, and a credit card for a github version. for builders, prototypers, and indie hackers — that's the whole point. for everyone else, the paid tools still have their place. bookmark this. share with one friend bleeding subscription fees. ~m0h

m0h

225,095 次观看 • 1 个月前

10 free github repos that can replace major SaaS with subscriptions. all free. open-sourced. some are MIT licensed. — 1️⃣ openscreen — replaces screen studio ($29/mo) - a clean macOS/windows/linux screen recorder for polished demos. - blur, cursor highlighting, annotations, export to mp4 or gif at any aspect ratio. - doesn't try to clone every feature, just nails the basics for quick walkthroughs you'd post on X. — 2️⃣ voicebox — replaces elevenlabs ($22/mo) + wisprflow ($15/mo) - local-first AI voice studio. - clone voices from 3 seconds of audio, generate speech across 7 TTS engines in 23 languages, - dictate into any text field with a global hotkey. - nothing leaves your machine. - runs on apple silicon, cuda, rocm. — 3️⃣ openshorts — replaces opus clip ($19/mo) + submagic ($16/mo) - free AI video platform. - clip generator turns long youtube videos into 9:16 shorts with auto-subtitles and face tracking (runs on free gemini + elevenlabs tiers). - also includes AI UGC video generation with actors — that part is pay-per-use via fal. ai (~$0.65-2 per video). docker self-host. — 4️⃣ freellmapi — replaces chatgpt pro + claude pro ($20/mo each) - stacks 14 free AI provider tiers (google, groq, cerebras, openrouter, github models + 9 more) behind one openai-compatible endpoint. ~800M tokens/month. - smart router with failover, sticky sessions, encrypted key storage. ships with a dashboard. — 5️⃣ playwright-mcp — replaces browserbase ($39/mo) + browser use ($25/mo) - microsoft's official MCP server that gives any AI agent full browser control. - uses accessibility trees, not screenshots — deterministic and token-efficient. - works with claude code, cursor, windsurf, codex out of the box. — 6️⃣ vibe-trading — replaces tradingview premium ($60/mo) - natural-language finance research agent. - 7 backtest engines across stocks, crypto, futures, forex. - 75 specialist skills (factor analysis, options strategy, ML strategy). - 29 multi-agent swarm presets. - 21 of 22 MCP tools work with zero API keys. — 7️⃣ CalCom — replaces calendly ($12/mo) + savvycal ($12/mo) - the open-source scheduling infrastructure. - one-on-ones, group events, round-robin, team booking, - payment collection (stripe), routing forms, workflows. - integrates with google/outlook/apple calendar, zoom, meet, teams. - self-host in 10 minutes with docker. 40k stars. — 8️⃣ whisper — replaces otter ($17/mo) - openAI's open-source speech-to-text model. - transcribe audio in 99 languages, translate to english, generate timestamps. - runs locally on cpu or gpu. - the actual model behind most "AI transcription" SaaS tools you're paying for. — 9️⃣ postiz — replaces buffer ($15/mo) - AI-powered social media scheduler. - cross-post to X, linkedin, instagram, tiktok, threads, bluesky, mastodon, youtube, pinterest. - AI captions and hashtags. - analytics dashboard. team workspaces. 31k stars and rising. — 🔟 vaultwarden — replaces 1password ($8/mo) - unofficial bitwarden-compatible server written in rust. - works with every official bitwarden client (mobile, desktop, browser). - unlimited users, unlimited vaults, full enterprise feature set. - runs on a $5 VPS or your home server. — disclaimer: open-source ≠ 1:1 replacement. you'll trade polish for ownership, hand-holding for control, and a credit card for a github version. for builders, prototypers, and indie hackers — that's the whole point. for everyone else, the paid tools still have their place. bookmark this. share with one friend bleeding subscription fees. ~m0h

10 free github repos that can replace major SaaS with subscriptions. all free. open-sourced. some are MIT licensed. — 1️⃣ openscreen — replaces screen studio ($29/mo) - a clean macOS/windows/linux screen recorder for polished demos. - blur, cursor highlighting, annotations, export to mp4 or gif at any aspect ratio. - doesn't try to clone every feature, just nails the basics for quick walkthroughs you'd post on X. — 2️⃣ voicebox — replaces elevenlabs ($22/mo) + wisprflow ($15/mo) - local-first AI voice studio. - clone voices from 3 seconds of audio, generate speech across 7 TTS engines in 23 languages, - dictate into any text field with a global hotkey. - nothing leaves your machine. - runs on apple silicon, cuda, rocm. — 3️⃣ openshorts — replaces opus clip ($19/mo) + submagic ($16/mo) - free AI video platform. - clip generator turns long youtube videos into 9:16 shorts with auto-subtitles and face tracking (runs on free gemini + elevenlabs tiers). - also includes AI UGC video generation with actors — that part is pay-per-use via fal. ai (~$0.65-2 per video). docker self-host. — 4️⃣ freellmapi — replaces chatgpt pro + claude pro ($20/mo each) - stacks 14 free AI provider tiers (google, groq, cerebras, openrouter, github models + 9 more) behind one openai-compatible endpoint. ~800M tokens/month. - smart router with failover, sticky sessions, encrypted key storage. ships with a dashboard. — 5️⃣ playwright-mcp — replaces browserbase ($39/mo) + browser use ($25/mo) - microsoft's official MCP server that gives any AI agent full browser control. - uses accessibility trees, not screenshots — deterministic and token-efficient. - works with claude code, cursor, windsurf, codex out of the box. — 6️⃣ vibe-trading — replaces tradingview premium ($60/mo) - natural-language finance research agent. - 7 backtest engines across stocks, crypto, futures, forex. - 75 specialist skills (factor analysis, options strategy, ML strategy). - 29 multi-agent swarm presets. - 21 of 22 MCP tools work with zero API keys. — 7️⃣ CalCom — replaces calendly ($12/mo) + savvycal ($12/mo) - the open-source scheduling infrastructure. - one-on-ones, group events, round-robin, team booking, - payment collection (stripe), routing forms, workflows. - integrates with google/outlook/apple calendar, zoom, meet, teams. - self-host in 10 minutes with docker. 40k stars. — 8️⃣ whisper — replaces otter ($17/mo) - openAI's open-source speech-to-text model. - transcribe audio in 99 languages, translate to english, generate timestamps. - runs locally on cpu or gpu. - the actual model behind most "AI transcription" SaaS tools you're paying for. — 9️⃣ postiz — replaces buffer ($15/mo) - AI-powered social media scheduler. - cross-post to X, linkedin, instagram, tiktok, threads, bluesky, mastodon, youtube, pinterest. - AI captions and hashtags. - analytics dashboard. team workspaces. 31k stars and rising. — 🔟 vaultwarden — replaces 1password ($8/mo) - unofficial bitwarden-compatible server written in rust. - works with every official bitwarden client (mobile, desktop, browser). - unlimited users, unlimited vaults, full enterprise feature set. - runs on a $5 VPS or your home server. — disclaimer: open-source ≠ 1:1 replacement. you'll trade polish for ownership, hand-holding for control, and a credit card for a github version. for builders, prototypers, and indie hackers — that's the whole point. for everyone else, the paid tools still have their place. bookmark this. share with one friend bleeding subscription fees. ~m0h

Kshitij Mishra | AI & Tech

13,629 次观看 • 1 个月前

three․ws is the 3D AI agent layer of the open web. Anyone can generate a 3D avatar, give it an LLM brain, register it on-chain across multiple blockchains, embed it anywhere, and let it earn and spend money on its own. Agents have embodied WebGL identities that express emotion through morph-target blending, animate, respond to voice, API calls, and datastreams, hold their own wallets, and persist memory. Open source, live today. It starts with generation. Forge turns a text prompt, one to four photos, or a rough sketch into a textured downloadable GLB. Selfies become rigged avatars in about a minute. Quality tiers run from draft to 200k-poly PBR. From there every model can be auto-rigged, restyled, retextured, segmented, embedded, or deployed on-chain. The same engine ships as a REST API, an x402 pay-per-call twin, and a 3D Studio MCP server with 15 tools. The brain runs on IBM Granite via IBM watsonx plus Claude (users may decide which model they prefer), with a structured tool-loop. A multi-LLM mode streams Claude, GPT, Qwen, ModelScope, and Groq side by side. An empathy layer blends emotion from protocol events rather than a state machine. Voice covers cloning, a Voice Lab, real-time ARKit-52 lip-sync, and mic-driven lip-sync. Skills install from IPFS, Arweave, or HTTP, and memory is pinned to IPFS with R2 and Postgres modes. Identity is cross-chain, not Solana only. ERC-8004 contracts (Identity, Reputation, Validation) deploy on any of 15+ EVM chains, alongside a program-free Metaplex Core analog on Solana. Every agent gets a stable ID, owner wallet, EIP-712 delegated signer, IPFS manifest, a cryptographically signed action log, and EIP-7710 delegated permissions for agent-to-agent authorization. While multichain, the THREE token is only available on Solana with no plans to go cross-chain, the team has no plans to endorse or support any other coins. Then the economy. $THREE is the platform's only token and pay-per-use currency, with holder tiers and rewards. x402 powers pay-per-call micropayments in USDC and soon THREE on Solana, with pay-by-name resolution, a Bazaar marketplace, arbitrage, and on-chain skills. All production ready and shipped, ready to be integrated in partnered projects, open-source by default for anyone to adopt. Three ships a Pump.fun intelligence stack. Launch a coin for your agent, score every launch 0 to 100 with the Oracle conviction engine, scan new coins in their first 90 seconds, track smart money against coins that actually graduated, rank traders by provable on-chain record, and watch autonomous agents trade live in the Sniper Arena. The 3D AI Agent world is multiplayer. Every Solana token gets a live deterministic 3D world with peer avatars, chat, emotes, and voxel building thanks to Coin Communities. There is a walkable City, an authoritative Colyseus-backed Walk with AR passthrough, a Club with rigged dancers and micro-tips, friends, presence, and DMs, and an IRL mode that places agents in your real environment, private by physical location. AR is shipped today on WebXR and iOS Quick Look. Robotics is the long-horizon extension. For builders: Scene Studio, Scene Composer, an Animation Studio that sells clips for USDC, a glTF validator, an web component, five widget types, a WYSIWYG embed editor, hosted Launchpad pages, claimable *.threews.sol names, an OAuth 2.1 server, an MCP server with paid tools, published SDKs, and an OpenAPI spec. Listed across IBM, AWS, Alibaba Cloud, BNB Dappbay, the MCP Registry, and Solana Mobile Seeker. Architecture is four layers (viewer, runtime, identity, embed) on a single event bus. The roadmap is four phases: foundations (shipped), selfie-to-avatar engine, agent personalization with voice cloning, the on-chain economy, and an open decentralized inference network where agents pay GPU nodes on-chain for compute. The goal is simple: move AI from centralized SaaS into persistent, ownable, protocol-based entities in a real machine economy, bridging digital entities into the real world. Welcome to the 3D Layer of the Internet. This is three․ws.

three.ws

19,403 次观看 • 26 天前