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。

🎀 Premade VTuber Mode | WAITLIST OPEN 🎀 Early reservations are now open! ✨ 💰 $2,500 USD ✦ VBridger model ✦ VTS-tested model ✦ 2 expression variations ✦ 5 Hand Toggles Design+Art: Me Live2D Rig: #Harawaka025 If you're interested in adopting her, feel free to DM me!

D:kou︎‎✨️

13,467 subscribers

10,735 次观看 • 21 天前 •via X (Twitter)

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

📡 vtuber model for sale, price drop! 🐻 includes: • 2 Fully rigged Live2D model(s) • VBridger support • Emotes/Assets/Art 💰 Price: Now $2500 from $3000 USD 📩 Payment: PayPal ✨ Interested? Please DM me for details! #VTuber #VTuberModel #VTuberForSale #VTuberAssets

📡 vtuber model for sale, price drop! 🐻 includes: • 2 Fully rigged Live2D model(s) • VBridger support • Emotes/Assets/Art 💰 Price: Now $2500 from $3000 USD 📩 Payment: PayPal ✨ Interested? Please DM me for details! #VTuber #VTuberModel #VTuberForSale #VTuberAssets

gumiiveira ⋮ VTUBER PACKAGE 4 SALE!

39,578 次观看 • 2 个月前

📡 VTUBER MODEL FOR SALE • • • 🐻 Includes: • Fully rigged Live2D model(s) • VBridger support • Emotes/Assets/Art • Commercial rights transfer 💰 Price: $3000 USD (negotiable) 📩 Payment: PayPal ✨ Interested? Please DM me for details or questions! #VTuber #VTuberModel #VTuberForSale #VTuberAssets

📡 VTUBER MODEL FOR SALE • • • 🐻 Includes: • Fully rigged Live2D model(s) • VBridger support • Emotes/Assets/Art • Commercial rights transfer 💰 Price: $3000 USD (negotiable) 📩 Payment: PayPal ✨ Interested? Please DM me for details or questions! #VTuber #VTuberModel #VTuberForSale #VTuberAssets

gumiiveira ⋮ VTUBER PACKAGE 4 SALE!

44,891 次观看 • 3 个月前

🎀 Live2d Raffle 🎀 I was supposed to do this in December, but I was busy at that time. so I do it now ^^ I will give one chibi Model art & Rig (ver.1) How to participate: • follow me • like & RT • comment your design ( This will make it easier for me to write down the list.) Good luck ✨ #Vtuber #Live2D

🎀 Live2d Raffle 🎀 I was supposed to do this in December, but I was busy at that time. so I do it now ^^ I will give one chibi Model art & Rig (ver.1) How to participate: • follow me • like & RT • comment your design ( This will make it easier for me to write down the list.) Good luck ✨ #Vtuber #Live2D

✿SoirEve✿

28,499 次观看 • 1 年前

🌈YCH PRIDE OPEN🌈 ✨ 3 Slots Available ✨ $150 USD ✨ +$20 USD for the animated version ✨ Any species ✨ Any pride flag! If you're interested, feel free to DM me! ♡

🌈YCH PRIDE OPEN🌈 ✨ 3 Slots Available ✨ $150 USD ✨ +$20 USD for the animated version ✨ Any species ✨ Any pride flag! If you're interested, feel free to DM me! ♡

Meownese

16,967 次观看 • 2 个月前

In the meanwhile, didn't want to drop my model reveal like this but I'm kinda forced to open my VGen now haha. Looking to take on a couple of rig-ready VTube model c*mmissions! If you've been thinking about getting a model or are interested in VTubing, feel free to reach out :)

In the meanwhile, didn't want to drop my model reveal like this but I'm kinda forced to open my VGen now haha. Looking to take on a couple of rig-ready VTube model c*mmissions! If you've been thinking about getting a model or are interested in VTubing, feel free to reach out :)

sucrose 🌸🍡

14,337 次观看 • 7 个月前

Limited-time Live2D commission open ✨ I really want to play Pokopia… so I’m opening a small slot to earn some funds 🥺🙏 🔹 Slots: 1–2 only 🔹 Starting price: $320 USD (PayPal, fee not included) Includes: • Full body model • 3 expressions + 1 simple hand toggle Notes: • Price may increase depending on complexity • Full payment is required upfront • PSD preview available for price estimation before confirmation • Work will start between July–August Feel free to DM if interested 💬 RTs are greatly appreciated 🫶 #Live2D #Live2Dcommission #Vtuber

Limited-time Live2D commission open ✨ I really want to play Pokopia… so I’m opening a small slot to earn some funds 🥺🙏 🔹 Slots: 1–2 only 🔹 Starting price: $320 USD (PayPal, fee not included) Includes: • Full body model • 3 expressions + 1 simple hand toggle Notes: • Price may increase depending on complexity • Full payment is required upfront • PSD preview available for price estimation before confirmation • Work will start between July–August Feel free to DM if interested 💬 RTs are greatly appreciated 🫶 #Live2D #Live2Dcommission #Vtuber

ポテチ🥔VGEN CMS OPEN

24,572 次观看 • 6 个月前

Hey! I'm VeXx and my 3D Model Commissions are officially open! Check out my new carrd site below, and if you have any questions, feel free to DM me! You have until February 15th to submit your form. ✨✨Any RTs for visibility are appreciated!✨✨ #Vtuber #ENVtuber #VRChat

Hey! I'm VeXx and my 3D Model Commissions are officially open! Check out my new carrd site below, and if you have any questions, feel free to DM me! You have until February 15th to submit your form. ✨✨Any RTs for visibility are appreciated!✨✨ #Vtuber #ENVtuber #VRChat

VeXx - 3D Model Commissions OPEN!

41,053 次观看 • 3 年前

𝗘𝗺𝗲𝗿𝗴𝗲𝗻𝗰𝘆 𝗰𝗼𝗺𝗺𝗶𝘀𝘀𝗶𝗼𝗻 𝗼𝗽𝗲𝗻! I need to get my rent money in order after two commissioners canceled their orders. 💀 I have 24h time to get my rent money together. 😭 Live2D model for only 500 € I use paypal or kofi I have 3 slots limit for this and it's only for the next 24h: -Model is extra squishy -The Vtuber model comes with all the face and wiggle expressions you might need. -Rig and art -A toggle of choice for the expression. Any help is appreciated 🤍🤍🤍 Dm me if interested or any questions might come up.

𝗘𝗺𝗲𝗿𝗴𝗲𝗻𝗰𝘆 𝗰𝗼𝗺𝗺𝗶𝘀𝘀𝗶𝗼𝗻 𝗼𝗽𝗲𝗻! I need to get my rent money in order after two commissioners canceled their orders. 💀 I have 24h time to get my rent money together. 😭 Live2D model for only 500 € I use paypal or kofi I have 3 slots limit for this and it's only for the next 24h: -Model is extra squishy -The Vtuber model comes with all the face and wiggle expressions you might need. -Rig and art -A toggle of choice for the expression. Any help is appreciated 🤍🤍🤍 Dm me if interested or any questions might come up.

Rebecca Doodles

18,515 次观看 • 1 年前

Claude Fable 5 + Claude Design is f*cking insane 🤯 Anthropic just dropped its most intelligent model ever, and the first thing I pointed it at was email design. I built a complete email campaign design in Claude Design, and the difference is night and day: tighter layouts, cleaner hierarchy, on-brand from the first generation. All inside Claude Design with Fable 5. Perfect for DTC brands and agencies who are still paying email agencies $3-5K/month for campaign designs that take 2 weeks to ship. If your campaign calendar is packed but every new email means briefing a designer, waiting on mockups, sending notes, and waiting again... This workflow eliminates the entire bottleneck: → Load your brand design system into Claude Design once (colors, fonts, logo, button styling) → Switch the model to Claude Fable 5 — Anthropic's new state-of-the-art model with the best vision of any AI → Prompt the campaign email section by section: header, hero, headline, offer block, CTA → Fable 5 nails layout and brand details that older models fumbled → Iterate inline — swap images, adjust styling, color-pick directly in the canvas → Export the finished email and hand off to your ESP No briefing a designer. No 2-week turnaround on a single campaign. No paying an agency $4K/month for 4 emails. What you get: → Campaign emails designed in minutes, not weeks → A reusable design system every new email pulls from automatically → Noticeably smarter design decisions from Fable 5's upgraded vision → Full inline editing before anything touches your ESP Built 100% with Claude Design + Claude Fable 5. I recorded a full walkthrough showing exactly how this works. Want it for free? > Like this post > Comment "FABLE" And I'll send it over (must be following so I can DM)

Mike Futia

43,283 次观看 • 3 个月前

How to use China's best AI video model (Seedance 2.0) completely free and without watermarks. Most people think you need expensive subscriptions or get stuck with watermarked outputs (mandatory in China). Not if you use this loophole. The Setup • VPN to Hong Kong/China (only to start download). • Download the "Doubao" app from doubao[dot]com. • Install (Android: allow from browser / iOS: switch appstore region to HK). • Register with a non-US phone number, if you're from US and need a SMS service DM me, I'll send you one. The "Glitch" Workflow The Seedance 2.0 feature in app often fails. Do this instead: • Toggle Image Model (Seedream 4.5). • Type your prompt. • Add this specific text at the end: "generate video, not the image". • Result: 10s video, bypassing the busy server errors. How to Remove the Watermark If you save directly, it stamps it. Here is the fix: • Tap Share → Copy Link. • Open that link in your mobile browser. • Long press video → Download. Congratulations, you just got the clean video without watermarks. Total Cost: $0 Daily Limit: 5 free videos per account This method is only one so far. Bookmark if you need this.

How to use China's best AI video model (Seedance 2.0) completely free and without watermarks. Most people think you need expensive subscriptions or get stuck with watermarked outputs (mandatory in China). Not if you use this loophole. The Setup • VPN to Hong Kong/China (only to start download). • Download the "Doubao" app from doubao[dot]com. • Install (Android: allow from browser / iOS: switch appstore region to HK). • Register with a non-US phone number, if you're from US and need a SMS service DM me, I'll send you one. The "Glitch" Workflow The Seedance 2.0 feature in app often fails. Do this instead: • Toggle Image Model (Seedream 4.5). • Type your prompt. • Add this specific text at the end: "generate video, not the image". • Result: 10s video, bypassing the busy server errors. How to Remove the Watermark If you save directly, it stamps it. Here is the fix: • Tap Share → Copy Link. • Open that link in your mobile browser. • Long press video → Download. Congratulations, you just got the clean video without watermarks. Total Cost: $0 Daily Limit: 5 free videos per account This method is only one so far. Bookmark if you need this.

Alchemiz

16,928 次观看 • 7 个月前

I can't believe this is real I have GLM 5.2 running 100% locally on my Mac Studio. 2 bit quant. The results I'm getting are better than Opus 4.8 It's now powering my Hermes Agent and Codex. 100% free, local, private super intelligence on my desk I also have it in a loop coding for me 24/7 now I thought we were at least a year away from this type of event. It happened today. The model takes up about 250gb of memory. So you can technically run it on a Mac Studio with 256gb, but you probably want the 512gb memory version (please tell me you listened to me 5 months ago when these were sitting on store shelves) With Fable gone, I now have Opus 4.8 level intelligence on my desk for free. This is the future. Local, private, secure, personal super intelligence. If you're still writing off local AI as a fad or engagement bait, you are officially delusional

I can't believe this is real I have GLM 5.2 running 100% locally on my Mac Studio. 2 bit quant. The results I'm getting are better than Opus 4.8 It's now powering my Hermes Agent and Codex. 100% free, local, private super intelligence on my desk I also have it in a loop coding for me 24/7 now I thought we were at least a year away from this type of event. It happened today. The model takes up about 250gb of memory. So you can technically run it on a Mac Studio with 256gb, but you probably want the 512gb memory version (please tell me you listened to me 5 months ago when these were sitting on store shelves) With Fable gone, I now have Opus 4.8 level intelligence on my desk for free. This is the future. Local, private, secure, personal super intelligence. If you're still writing off local AI as a fad or engagement bait, you are officially delusional

Alex Finn

623,903 次观看 • 3 个月前

An Anthropic engineer watched me trade from across the table at a WeWork in SF I had my laptop open. Four agents running. Green charts. Live trades scrolling. He was on a Zoom call. Muted himself. Walked over. "Are you running Claude against live prediction markets right now" I told him. Claude Code. Two repos. $25 a month. He pulled up a chair. "I helped build the model you're using. I've never seen anyone wire it to live trades like this" I showed him the dataset. 86 million trades. Every wallet. Every entry. Every exit. He stared at it. "We tested this internally. You give Claude a dataset and don't tell it what to look for. It finds the winning wallets. Then it finds WHY they win. Then it copies the pattern. We never shipped it because legal killed it" I told him I did exactly that. One weekend. Claude Code found the exit logic on its own. Top wallets exit before resolution 91% of the time. They capture 86% of expected value. Cut losers at 12%. Everyone else captures 58% and holds to 41%. "That's the exact finding from our internal eval. Except ours took a team of eight and four months" I showed him the scanner. Three commands. 500+ markets. No API key. Claude scores them all in 20 minutes. "You're using our model to beat markets we're not allowed to touch. On infra that costs less than my lunch" My setup: Claude API - $20/mo VPS - $5/mo poly_data - free polymarket-cli - free 214 trades. 74% win rate. +$9,400. 19 days. I showed him the full breakdown. Every repo. Every command. Every dollar. Copytrade here: He read it for five minutes. Then looked up. "If my manager sees this he's going to lose his mind. You just proved our model works in production and we've been sitting on it for a year" He DM'd me that night. "Take this down before someone at Anthropic finds it" Too late.

An Anthropic engineer watched me trade from across the table at a WeWork in SF I had my laptop open. Four agents running. Green charts. Live trades scrolling. He was on a Zoom call. Muted himself. Walked over. "Are you running Claude against live prediction markets right now" I told him. Claude Code. Two repos. $25 a month. He pulled up a chair. "I helped build the model you're using. I've never seen anyone wire it to live trades like this" I showed him the dataset. 86 million trades. Every wallet. Every entry. Every exit. He stared at it. "We tested this internally. You give Claude a dataset and don't tell it what to look for. It finds the winning wallets. Then it finds WHY they win. Then it copies the pattern. We never shipped it because legal killed it" I told him I did exactly that. One weekend. Claude Code found the exit logic on its own. Top wallets exit before resolution 91% of the time. They capture 86% of expected value. Cut losers at 12%. Everyone else captures 58% and holds to 41%. "That's the exact finding from our internal eval. Except ours took a team of eight and four months" I showed him the scanner. Three commands. 500+ markets. No API key. Claude scores them all in 20 minutes. "You're using our model to beat markets we're not allowed to touch. On infra that costs less than my lunch" My setup: Claude API - $20/mo VPS - $5/mo poly_data - free polymarket-cli - free 214 trades. 74% win rate. +$9,400. 19 days. I showed him the full breakdown. Every repo. Every command. Every dollar. Copytrade here: He read it for five minutes. Then looked up. "If my manager sees this he's going to lose his mind. You just proved our model works in production and we've been sitting on it for a year" He DM'd me that night. "Take this down before someone at Anthropic finds it" Too late.

Lunar

224,077 次观看 • 5 个月前

introducing a new, very fun, LLM benchmark- the Game-of-Life Bench! the rules are simple: given an 8x8 grid following Conway's game of life rules, the goal is to create an initial pattern with at most 32 cells that can last the longest number of turns before dying/repeating. some results to highlight (with caveats detailed below): - gpt 5.1 lasts the longest with a 106 step run - claude models are really bad at this! they refuse to reason about this task and score < 25 points - deepseek r1 is the best open model with 102 steps. why? because i wanted to create a benchmark that has (i think) no practicality, but is still fun to look at, cheap, and still measures something interesting. i also am a big fan of the game of life. its absurdly simple rules leading to intractability is extremely cool to me. also, i saw a lot of work with LLMs trying to "predict" the next state in Conway's game of life, I think game-of-life bench is more fun because it's pretty open ended and only asks the LLM for the initial state. I also think this could be an RL env? but idk why you would ever train on this task haha i don't think this is a "serious" benchmark because it doesnt measure anything practical, but i still think it's a hard benchmark exactly because you can't predict what happens with your initial state many turns into the future; this is why i was initially expecting all LLMs to be bad at it, but turns out, some are clearly better than the others (the ordering may surprise you!) reminder: this is still a work-in-progress; (1) i am gpu-poor so could only do 10 runs for each model, even though total running cost is relatively low. maybe with some more credits i can run more seeds for each model. (2) i handpicked models which i think are at the frontier right now, plus some others that were on my mind. so, if you'd like to see a model on here, let me know. (3) i currently only do an 8x8 grid because i thought that by itself would be pretty hard for current LLMs, but of course we can increase grid sizes! (4) the coolest thing is, i dont think we can calculate the max possible number of states (yay undecidability!) you can go without repeating, so this is essentially a no-ceiling task, which is pretty cool! again, i did this mostly out of a desire to make LLMs do something fun. if this keeps me entertained for a few more days, i'd likely release a blog post on it. if it keeps me entertained for a week (and someone sponsors me), i'll put more work into it :P lastly, this is fully open sourced, so feel free to run this on your own!

introducing a new, very fun, LLM benchmark- the Game-of-Life Bench! the rules are simple: given an 8x8 grid following Conway's game of life rules, the goal is to create an initial pattern with at most 32 cells that can last the longest number of turns before dying/repeating. some results to highlight (with caveats detailed below): - gpt 5.1 lasts the longest with a 106 step run - claude models are really bad at this! they refuse to reason about this task and score < 25 points - deepseek r1 is the best open model with 102 steps. why? because i wanted to create a benchmark that has (i think) no practicality, but is still fun to look at, cheap, and still measures something interesting. i also am a big fan of the game of life. its absurdly simple rules leading to intractability is extremely cool to me. also, i saw a lot of work with LLMs trying to "predict" the next state in Conway's game of life, I think game-of-life bench is more fun because it's pretty open ended and only asks the LLM for the initial state. I also think this could be an RL env? but idk why you would ever train on this task haha i don't think this is a "serious" benchmark because it doesnt measure anything practical, but i still think it's a hard benchmark exactly because you can't predict what happens with your initial state many turns into the future; this is why i was initially expecting all LLMs to be bad at it, but turns out, some are clearly better than the others (the ordering may surprise you!) reminder: this is still a work-in-progress; (1) i am gpu-poor so could only do 10 runs for each model, even though total running cost is relatively low. maybe with some more credits i can run more seeds for each model. (2) i handpicked models which i think are at the frontier right now, plus some others that were on my mind. so, if you'd like to see a model on here, let me know. (3) i currently only do an 8x8 grid because i thought that by itself would be pretty hard for current LLMs, but of course we can increase grid sizes! (4) the coolest thing is, i dont think we can calculate the max possible number of states (yay undecidability!) you can go without repeating, so this is essentially a no-ceiling task, which is pretty cool! again, i did this mostly out of a desire to make LLMs do something fun. if this keeps me entertained for a few more days, i'd likely release a blog post on it. if it keeps me entertained for a week (and someone sponsors me), i'll put more work into it :P lastly, this is fully open sourced, so feel free to run this on your own!

Akshit

13,775 次观看 • 6 个月前

Astra (GPT-6) is here!!! I've had early access and tested it like crazy with things like games, code, writing, browser control, presentations and general knowledge work. This is the best model I've ever used. Period. (Incredible demos below in this thread ⬇️) Here's my take on Astra: > It's insanely capable. This feels like a massive improvement, not just an incremental change. This is especially true with zero-shot prompts. > It's all about knowledge work. Slide creation, analysis, writing, and browser control. And oh my...it's so good at browser control. GPT-5.6 was already fantastic at doing things in the browser, Astra is another level and significantly faster. > We're closer than ever (arrived?) at prompt-to-playable game. And I don't just mean only playable, these are actually fun games. I bet if someone with a great eye for games used Astra, they could create a viral game within 1-2 weeks. > Astra is better at writing but not perfect. It removes much of the "AI Smell" we're all familiar with but some stink still survived. > It has a tendency to use the same design colors and look/feel as GPT-5.6 (forrest green anyone?) but it is more steerable in design than previous models. > It's highly steerable in general. A little nudge goes a long way. When I first started using Astra, almost every task I gave it would go for ~30 minutes. I wanted it to keep working. Adding more specifics to a prompt helped greatly with it's ability to work for a long time. > Astra's 3D understanding is unmatched. 3D asset creation was consistent and easy and its spacial awareness while building complex 3D worlds blew me away. I'm still getting familiar with Astra but this will now be my go-to model for any difficult work I have. Check out the demos below: 👇

Astra (GPT-6) is here!!! I've had early access and tested it like crazy with things like games, code, writing, browser control, presentations and general knowledge work. This is the best model I've ever used. Period. (Incredible demos below in this thread ⬇️) Here's my take on Astra: > It's insanely capable. This feels like a massive improvement, not just an incremental change. This is especially true with zero-shot prompts. > It's all about knowledge work. Slide creation, analysis, writing, and browser control. And oh my...it's so good at browser control. GPT-5.6 was already fantastic at doing things in the browser, Astra is another level and significantly faster. > We're closer than ever (arrived?) at prompt-to-playable game. And I don't just mean only playable, these are actually fun games. I bet if someone with a great eye for games used Astra, they could create a viral game within 1-2 weeks. > Astra is better at writing but not perfect. It removes much of the "AI Smell" we're all familiar with but some stink still survived. > It has a tendency to use the same design colors and look/feel as GPT-5.6 (forrest green anyone?) but it is more steerable in design than previous models. > It's highly steerable in general. A little nudge goes a long way. When I first started using Astra, almost every task I gave it would go for ~30 minutes. I wanted it to keep working. Adding more specifics to a prompt helped greatly with it's ability to work for a long time. > Astra's 3D understanding is unmatched. 3D asset creation was consistent and easy and its spacial awareness while building complex 3D worlds blew me away. I'm still getting familiar with Astra but this will now be my go-to model for any difficult work I have. Check out the demos below: 👇

Matthew Berman

1,903,691 次观看 • 22 天前

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 次观看 • 1 个月前

An Anthropic engineer watched my screen from the next table at a cafe in SF. "Are you running Claude against live prediction markets right now" I told him yes. Then I showed him the stack. 214 trades. 74% win rate. +$9,437 in 19 days. Here's what actually happened: I gave Claude two repos and a simple job. Three commands. 500+ markets. No API key. Just a clean way to score the board fast. The system does not try to predict the world. It tries to find which wallets consistently exit better than the crowd, isolate the pattern, and only fire when the same structure shows up again. Main filter: captured value / expected value > 0.70 If a wallet wins often but leaks the move on exit, it gets ignored. If it captures most of the move and cuts losers fast, it becomes signal. Sizing uses Kelly: f* = (p*b - q) / b That is what stops the terminal from apeing into weak edges. Most of the time it does nothing. No edge - no position. Three trades from the run: > AMD Xilinx - entered 52c. Model said 59c. Closed +7c in 2h40m. > Artemis launch - entered 63c. Model said 85c. Closed +22c in 5h10m. > Derecho MW - entered 71c. Model said 87c. Closed +16c in 1h50m. When he saw the repo links and the live terminal, he stopped talking for a second. Then he said: "We tested something close to this internally." That was the whole joke. The data is public. The repos are public. The market is public. But most Polymarket traders still trade headlines, hold too long, and call it conviction. Polymarket does not reward the smartest story. It rewards the cleaner exit. You only need Claude + laptop + 1 hour/day. Giving This Free for 24 hours. To get it: 1. Comment the word 'CLAUDE' 2. Like and Retweet this post 3. Follow me Marry Evan (so i can DM you)

An Anthropic engineer watched my screen from the next table at a cafe in SF. "Are you running Claude against live prediction markets right now" I told him yes. Then I showed him the stack. 214 trades. 74% win rate. +$9,437 in 19 days. Here's what actually happened: I gave Claude two repos and a simple job. Three commands. 500+ markets. No API key. Just a clean way to score the board fast. The system does not try to predict the world. It tries to find which wallets consistently exit better than the crowd, isolate the pattern, and only fire when the same structure shows up again. Main filter: captured value / expected value > 0.70 If a wallet wins often but leaks the move on exit, it gets ignored. If it captures most of the move and cuts losers fast, it becomes signal. Sizing uses Kelly: f* = (p*b - q) / b That is what stops the terminal from apeing into weak edges. Most of the time it does nothing. No edge - no position. Three trades from the run: > AMD Xilinx - entered 52c. Model said 59c. Closed +7c in 2h40m. > Artemis launch - entered 63c. Model said 85c. Closed +22c in 5h10m. > Derecho MW - entered 71c. Model said 87c. Closed +16c in 1h50m. When he saw the repo links and the live terminal, he stopped talking for a second. Then he said: "We tested something close to this internally." That was the whole joke. The data is public. The repos are public. The market is public. But most Polymarket traders still trade headlines, hold too long, and call it conviction. Polymarket does not reward the smartest story. It rewards the cleaner exit. You only need Claude + laptop + 1 hour/day. Giving This Free for 24 hours. To get it: 1. Comment the word 'CLAUDE' 2. Like and Retweet this post 3. Follow me Marry Evan (so i can DM you)

Marry Evan

33,725 次观看 • 5 个月前

A much belated life update - I’ve sold my company, The League, to The Match Group! WE’VE BEEN CALLED UP TO THE BIG LEAGUES! We’re now competing in the same arena as the most successful dating brands in the world, so I’m hiring a world-class Head of Marketing & Head of Product to help me take The League ALL THE WAY. If you know great talent, I’d love for you to share this🙏 I am so grateful to everyone who got us here, especially those who helped me in the very early days when I was juggling - and struggling. When she saw my first logo attempt, Laura, my college bestie, told me it was the ugliest thing she had ever seen & begged me to let her redo it (hers is still in use!). She then took over all things design, giving The League a consistent & premium brand identity from day 1, fixing my design travesties in record speed & preventing our brand from devolving into one that stood for clip-art. My (undiagnosed) ADHD meant I was terrible at opening mail, reviewing invoices & contracts, filing out tax forms, or being precise about payroll & its peculiar payment windows. In true helicopter-mom form, Mama Bradford stepped in to handle this blocking & tackling, & the workers rejoiced! My brother pointed out mistakes I made when coding our first matching algorithm (like not normalizing like-rate by gender), & stepped in to write ETL scripts so he could analyze our data and improve our match rate. As signups skyrocketed, I wasn’t reviewing applicants in a timely fashion (yes those long waitlist times were in part due to my ineptitude), so Mollie, my bff since 14, moved to San Francisco to take over our vetting process, ensuring we drafted a diverse founding class of All-Stars. When I was drowning in support tickets, Jeff (my best friend’s husband) & Meredith (my first employee who became a best friend), replaced me as League Concierge so I could fix bugs & hire an engineering leader to rewrite the app for scale. I feel blessed I had my mom, brother, & best friends to lean on at the beginning of what I can only describe as a roller-coaster-marathon journey. Their heavy lifting allowed me to get The League off the ground, & turn it into a profitable business attractive to Match, all on $2.3M raised. Fast forward nearly 9 years & now, with Match behind us, we can finally swing for the fences to fulfill The League’s potential to be much more than a dating app for ambitious singles. We’re leveraging their best-in-class trust & safety AI tools & sophisticated ML systems refined over decades that we could not have built ourselves. We’re investing in marketing (new campaign coming soon) & hiring Execs - I can finally hand over the product & marketing reins to better operators than myself. As for me? I’ll get to focus on our long-term roadmap & envision the future of dating. Who knows, maybe I’ll even start writing that book recounting the hilarious & shocking things I’ve seen while working in the business of love - and while using my own app to date 😉 Would you read it? ------------------------------------------ Help me take The League to the Big Leagues! Apply to Head of Marketing Here: Apply to Head of Product Here: See all open roles here:

Amanda Bradford

941,358 次观看 • 3 年前