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。

NariLabs Dia-1.6B by Toby Kim now on MLX 🚀 Get started: > pip install -U mlx-audio Model Card 👇🏾

Prince Canuma

6,315 subscribers

24,057 次观看 • 1 年前 •via X (Twitter)

艺术科学技术教育

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

Qwen2.5-VL now on MLX 🚀 You can now run inference or train it locally on your Mac using MLX. > pip install -U mlx-vlm Note: Video support is coming shortly :) Model cards 👇🏾

Qwen2.5-VL now on MLX 🚀 You can now run inference or train it locally on your Mac using MLX. > pip install -U mlx-vlm Note: Video support is coming shortly :) Model cards 👇🏾

Prince Canuma

29,952 次观看 • 1 年前

LLaVA Llama-3 and Phi-3 now on MLX 🎉🚀 You can now run inference locally on your Mac. pip install -U mlx-vlm I’m getting ~50 tokens on a M3 Max. Model cards 👇🏾

LLaVA Llama-3 and Phi-3 now on MLX 🎉🚀 You can now run inference locally on your Mac. pip install -U mlx-vlm I’m getting ~50 tokens on a M3 Max. Model cards 👇🏾

Prince Canuma

74,448 次观看 • 2 年前

Chatterbox Turbo by Resemble AI now on MLX 🚀🎉 You can now run it locally on your Mac and it supports voice cloning and emotion control. I'm getting 3.8x faster than real-time. > pip install -U mlx-audio Model collection 👇🏽

Chatterbox Turbo by Resemble AI now on MLX 🚀🎉 You can now run it locally on your Mac and it supports voice cloning and emotion control. I'm getting 3.8x faster than real-time. > pip install -U mlx-audio Model collection 👇🏽

Prince Canuma

24,953 次观看 • 7 个月前

Sam 3 by Facebook now on MLX 🚀 Here is a realtime object tracking running on M3 Max 96GB.

Sam 3 by Facebook now on MLX 🚀 Here is a realtime object tracking running on M3 Max 96GB.

Prince Canuma

180,321 次观看 • 3 个月前

$Really impressed by quality of Marvis-AI made by Prince Canuma & Lucas Newman, it's only 250M and uses less than 500MB in 4-bit 🤯 You only need 2 lines of code to run on macOS: pip install -U mlx-audio python -m mlx_audio.tts.generate --model Marvis-AI/marvis-tts-250m-v0.1 --stream \ --text "Hello there!" No reason not to give it a try!$

Really impressed by quality of Marvis-AI made by Prince Canuma & Lucas Newman, it's only 250M and uses less than 500MB in 4-bit 🤯 You only need 2 lines of code to run on macOS: pip install -U mlx-audio python -m mlx_audio.tts.generate --model Marvis-AI/marvis-tts-250m-v0.1 --stream \ --text "Hello there!" No reason not to give it a try!

Adrien Grondin

11,780 次观看 • 11 个月前

Playing with Qwen3-TTS is and MLX-Audio locally on Mac Studio M3 Ultra 🔥 Amazing model by Qwen and great work by Prince Canuma bringing this magic to MLX! Tuning the voice with instructions feels like magic! Command and prompts to run it are in the video.

Playing with Qwen3-TTS is and MLX-Audio locally on Mac Studio M3 Ultra 🔥 Amazing model by Qwen and great work by Prince Canuma bringing this magic to MLX! Tuning the voice with instructions feels like magic! Command and prompts to run it are in the video.

Ivan Fioravanti ᯅ

35,720 次观看 • 6 个月前

Quick demo of LFM 2.5 VL 1.6B model by Liquid AI that I recently added to Locally AI - Local AI Chat Running locally on iPhone 17 Pro at ~90tk/s with MLX Small vision-language models are improving fast

Quick demo of LFM 2.5 VL 1.6B model by Liquid AI that I recently added to Locally AI - Local AI Chat Running locally on iPhone 17 Pro at ~90tk/s with MLX Small vision-language models are improving fast

Adrien Grondin

32,556 次观看 • 6 个月前

Demo of 1-bit Bonsai 8B from PrismML running on-device on iPhone 17 Pro More than 40tk/s for a dense 8B model on iPhone, that’s a first Powered by Apple MLX and available now in Locally AI

Demo of 1-bit Bonsai 8B from PrismML running on-device on iPhone 17 Pro More than 40tk/s for a dense 8B model on iPhone, that’s a first Powered by Apple MLX and available now in Locally AI

Adrien Grondin

160,059 次观看 • 3 个月前

We (Claude and I) made a new package: voxmlx It's an MLX implementation of Mistral's Voxtral mini realtime speech recognition model. It supports streaming audio and runs pretty fast on a laptop. To use it, simply: `uvx voxmlx` Also I did not write a single line of code for this package. Every line of code was written by Claude Code. More on that in thread.

We (Claude and I) made a new package: voxmlx It's an MLX implementation of Mistral's Voxtral mini realtime speech recognition model. It supports streaming audio and runs pretty fast on a laptop. To use it, simply: `uvx voxmlx` Also I did not write a single line of code for this package. Every line of code was written by Claude Code. More on that in thread.

Awni Hannun

33,896 次观看 • 5 个月前

Perplexity users can now onramp crypto powered by the MoonPay Model Context Protocol (MCP) get started in seconds on Perplexity using our new Custom Connector to give your agents access to crypto here's a look at how it works:

Perplexity users can now onramp crypto powered by the MoonPay Model Context Protocol (MCP) get started in seconds on Perplexity using our new Custom Connector to give your agents access to crypto here's a look at how it works:

MoonPay 🟣

56,550 次观看 • 26 天前

1/ World models are getting popular in robotics 🤖✨ But there’s a big problem: most are slow and break physical consistency over long horizons. 2/ Today we’re releasing Interactive World Simulator: An action-conditioned world model that supports stable long-horizon interaction. 3/ Key result: ✅ 10+ minutes of interactive prediction ✅ 15 FPS ✅ on a single RTX 4090🔥 4/ Why this matters: it unlocks two critical robotics applications: 🚀 Scalable data generation for policy training 🧪 Faithful policy evaluation 5/ You can play with our world model NOW at NO git clone, NO pip install, NO python. Just click and play! NOTE ⚠️ ALL videos here are generated purely by our model in pixel space! They are **NOT** from a real camera More details coming 👇 (1/9) #Robotics #AI #MachineLearning #WorldModels #RobotLearning #ImitationLearning

1/ World models are getting popular in robotics 🤖✨ But there’s a big problem: most are slow and break physical consistency over long horizons. 2/ Today we’re releasing Interactive World Simulator: An action-conditioned world model that supports stable long-horizon interaction. 3/ Key result: ✅ 10+ minutes of interactive prediction ✅ 15 FPS ✅ on a single RTX 4090🔥 4/ Why this matters: it unlocks two critical robotics applications: 🚀 Scalable data generation for policy training 🧪 Faithful policy evaluation 5/ You can play with our world model NOW at NO git clone, NO pip install, NO python. Just click and play! NOTE ⚠️ ALL videos here are generated purely by our model in pixel space! They are NOT from a real camera More details coming 👇 (1/9) #Robotics #AI #MachineLearning #WorldModels #RobotLearning #ImitationLearning

Yixuan Wang

128,556 次观看 • 4 个月前

🚀 The battle cry echoes, Cyber Gladiators! 🚀 🔥Powered By BNB Chain, our game and the arena of staked fights have RISEN! 🔥 🎮Get your digits in formation, for the quest to earn mighty $CAT and $DOGE begins NOW! 💰 🌐 Available on ALL fronts: • 🌐 WebGL - LIVE - • 📱 Android - LIVE - • 🍏 iOS - LIVE (sans staked fights) - 🔥 The Cyber Arena awaits your valor, and the rewards are legendary! 🏆💫 🎁 For our early warriors, 200 $CAT await you! Install, enter the arena, and let the grind COMMENCE! ⚔️💥 🍏 While we encountered some unexpected challenges with the Apple gods (thanks to an uncanny resemblance between Elon Musk and Mark Zuckerberg) stay tuned for their verdict on iOS staked battles! 🤖⚡️ #CyberArena #Crypto #gaming #p2e

🚀 The battle cry echoes, Cyber Gladiators! 🚀 🔥Powered By BNB Chain, our game and the arena of staked fights have RISEN! 🔥 🎮Get your digits in formation, for the quest to earn mighty $CAT and $DOGE begins NOW! 💰 🌐 Available on ALL fronts: • 🌐 WebGL - LIVE - • 📱 Android - LIVE - • 🍏 iOS - LIVE (sans staked fights) - 🔥 The Cyber Arena awaits your valor, and the rewards are legendary! 🏆💫 🎁 For our early warriors, 200 $CAT await you! Install, enter the arena, and let the grind COMMENCE! ⚔️💥 🍏 While we encountered some unexpected challenges with the Apple gods (thanks to an uncanny resemblance between Elon Musk and Mark Zuckerberg) stay tuned for their verdict on iOS staked battles! 🤖⚡️ #CyberArena #Crypto #gaming #p2e

Cyber Arena AR/VR

24,627 次观看 • 2 年前

#ContestAlert 🚨 Make your Diwali a pay-tastic affair! 🎇 Participate in the #CheQItToWinIt contest and be one of the 7 lucky winners to get a Nothing Phone (2a) Plus!📱 Here’s how to get in on the festive fun: ✅Download the CheQ app ✅Pay your credit card bill on CheQ ✅Take a screenshot of your payment confirmation screen ✅Share the screenshot, tag CheQ and use #CheQItToWinIt. Make sure you’re following CheQ too! 🏆Every day, one lucky winner wins a Nothing Phone (2a) Plus! 🔊Rules ✅Contest duration: Oct 30 - Nov 5, 2024. ✅Only payment made within the contest period will be considered valid. ✅A minimum total transaction of ₹5000 is required to qualify ✅Multiple entries allowed with different payment screenshots. 📢Winner Announcement ✅All the eligible entries will be selected after verification of their payment by the CheQ Team. ✅All the winners will be announced on this page on November 11. 🔥Brownie points on more retweets! 👀 It’s that simple to make this Diwali truly fantastic! Participate now! 🚀 #CheQItToWinIt #ContestAlert #NothingPhone #GiveawayAlert #CreditCards #CheQ #Diwali #creditcardbillpayments

#ContestAlert 🚨 Make your Diwali a pay-tastic affair! 🎇 Participate in the #CheQItToWinIt contest and be one of the 7 lucky winners to get a Nothing Phone (2a) Plus!📱 Here’s how to get in on the festive fun: ✅Download the CheQ app ✅Pay your credit card bill on CheQ ✅Take a screenshot of your payment confirmation screen ✅Share the screenshot, tag CheQ and use #CheQItToWinIt. Make sure you’re following CheQ too! 🏆Every day, one lucky winner wins a Nothing Phone (2a) Plus! 🔊Rules ✅Contest duration: Oct 30 - Nov 5, 2024. ✅Only payment made within the contest period will be considered valid. ✅A minimum total transaction of ₹5000 is required to qualify ✅Multiple entries allowed with different payment screenshots. 📢Winner Announcement ✅All the eligible entries will be selected after verification of their payment by the CheQ Team. ✅All the winners will be announced on this page on November 11. 🔥Brownie points on more retweets! 👀 It’s that simple to make this Diwali truly fantastic! Participate now! 🚀 #CheQItToWinIt #ContestAlert #NothingPhone #GiveawayAlert #CreditCards #CheQ #Diwali #creditcardbillpayments

CheQ

248,463 次观看 • 1 年前

🪴 GT Protocol Monthly Recap: May 2026 May focused on launching advanced trading infrastructure, introducing AI risk-management tools, and shipping major platform upgrades. 🚀 Hyperliquid Vaults Live Run multiple algorithmic strategies on a single Hyperliquid Vault inside GT App. Enjoy automated execution, auto-rebalancing, and protocol-level security. You can find Vault trading on the Hyperliquid exchange account connection page in the Trade on Vault section. Try it in GT App 👉 🤖 AI Hedge Fund Experiment Live An experimental AI Hedge Fund powered by 5 independent LLM models is live on Hyperliquid. Each model manages $10,000 to test different AI trading personalities and allocation strategies. Discover it now here 👉 📈 Isolated Margin & AI Risk Tools Isolated Margin is live across GT App for precise risk management. Enhanced with AI-powered logic, it assists with dynamic asset monitoring and smarter strategy deployment. Try it in GT App 👉 🔥 Top Strategy Performance Top trader strategies like "lebakien" achieved over +141% profit this month. Users can explore metrics and follow the strategies of top traders directly in the marketplace. Explore Marketplace 👉 🛠 Key Product Updates ⚙️ Strategy Discovery: enhanced demo trading flows and top trader strategy integration. ⚙️ AI Strategy Chat: demoed a flow to create, launch, and test strategies via natural language chat. ⚙️ Advanced Execution: added manual safety orders for granular control over active positions. ⚙️ Testing & Validation: optimized historical data validation for more accurate strategy testing. ⚙️ Knowledge Hub: launched GT Protocol Learn and a new Knowledge Base for streamlined support. ⚙️ Performance: upgraded website structure and improved overall page responsiveness. Find all the latest GT App updates Here 👉 Discover guides, insights, and resources in Learn 👉 and Knowledge Base 👉 📰 GT Protocol AI Digests 4 new AI Digest issues (No.89–92) are live on Medium, covering AI-native hardware, data privacy, and the evolution of AI agents. Read More 👉 May brought institutional-grade AI strategy management closer to every user.

🪴 GT Protocol Monthly Recap: May 2026 May focused on launching advanced trading infrastructure, introducing AI risk-management tools, and shipping major platform upgrades. 🚀 Hyperliquid Vaults Live Run multiple algorithmic strategies on a single Hyperliquid Vault inside GT App. Enjoy automated execution, auto-rebalancing, and protocol-level security. You can find Vault trading on the Hyperliquid exchange account connection page in the Trade on Vault section. Try it in GT App 👉 🤖 AI Hedge Fund Experiment Live An experimental AI Hedge Fund powered by 5 independent LLM models is live on Hyperliquid. Each model manages $10,000 to test different AI trading personalities and allocation strategies. Discover it now here 👉 📈 Isolated Margin & AI Risk Tools Isolated Margin is live across GT App for precise risk management. Enhanced with AI-powered logic, it assists with dynamic asset monitoring and smarter strategy deployment. Try it in GT App 👉 🔥 Top Strategy Performance Top trader strategies like "lebakien" achieved over +141% profit this month. Users can explore metrics and follow the strategies of top traders directly in the marketplace. Explore Marketplace 👉 🛠 Key Product Updates ⚙️ Strategy Discovery: enhanced demo trading flows and top trader strategy integration. ⚙️ AI Strategy Chat: demoed a flow to create, launch, and test strategies via natural language chat. ⚙️ Advanced Execution: added manual safety orders for granular control over active positions. ⚙️ Testing & Validation: optimized historical data validation for more accurate strategy testing. ⚙️ Knowledge Hub: launched GT Protocol Learn and a new Knowledge Base for streamlined support. ⚙️ Performance: upgraded website structure and improved overall page responsiveness. Find all the latest GT App updates Here 👉 Discover guides, insights, and resources in Learn 👉 and Knowledge Base 👉 📰 GT Protocol AI Digests 4 new AI Digest issues (No.89–92) are live on Medium, covering AI-native hardware, data privacy, and the evolution of AI agents. Read More 👉 May brought institutional-grade AI strategy management closer to every user.

GT Protocol

32,774 次观看 • 1 个月前

🎉 The countdown to the New Year is on, and at XPMarket, we’re celebrating in style - with new features and rewards! 🚀✨ You can now grab your Season Pass⚡️ by burning XPM tokens - just head to the Season Pass tab’s main banner to get started. 🎟️ But hurry! The season ends on Jan 12, so it’s the perfect time to catch up. 🎁 Speaking of rewards, a 300⚡️ bonus has just been revealed, translating to a whopping 2k RDC tokens from the bullish family at Rich Ducks Crypto Club 🦆💰 On the tech front, XPMarket is now 3x faster and primed to handle even more users—your experience just leveled up! Plus, we’ve added some subtle but handy new features. 🛠️ Think you can spot them? Drop your guesses in the comments below! 👀👇 Oh, and don’t party too hard tonight… because tomorrow we’re dropping some very limited and special rewards that you won’t want to miss! 🎁 Let’s toast to a brighter, faster, and more rewarding 2024 together 🥂✨ #NewYear #XRP #XRPL #Reward #XRParmy

🎉 The countdown to the New Year is on, and at XPMarket, we’re celebrating in style - with new features and rewards! 🚀✨ You can now grab your Season Pass⚡️ by burning XPM tokens - just head to the Season Pass tab’s main banner to get started. 🎟️ But hurry! The season ends on Jan 12, so it’s the perfect time to catch up. 🎁 Speaking of rewards, a 300⚡️ bonus has just been revealed, translating to a whopping 2k RDC tokens from the bullish family at Rich Ducks Crypto Club 🦆💰 On the tech front, XPMarket is now 3x faster and primed to handle even more users—your experience just leveled up! Plus, we’ve added some subtle but handy new features. 🛠️ Think you can spot them? Drop your guesses in the comments below! 👀👇 Oh, and don’t party too hard tonight… because tomorrow we’re dropping some very limited and special rewards that you won’t want to miss! 🎁 Let’s toast to a brighter, faster, and more rewarding 2024 together 🥂✨ #NewYear #XRP #XRPL #Reward #XRParmy

xpmarket.com

16,142 次观看 • 1 年前

This is my "feel the AGI" moment: I used GPT-5.6 Sol to train my own autocorrect model that outperforms GPT-5.6 Sol (wtf??) I have no ML background. I have no idea what I'm doing. I just kept pushing Sol until it spat out a SOTA model. And I spent $0. The motivation: Years of talking to AI have made me terrible at typing. Rather than fix my skill issue, I decided to throw more AI at it. My idea was: instead of autocorrect that interrupts my flow, I want to type fast with mistakes and have AI clean it up after. I wanted the smallest local model possible, for speed, for battery life, for science! So I decided to train my own. Inspired by Andrej Karpathy’s autoresearch, I ran Codex /goal with this setup: pick an experiment, try it, record the results to a doc, throw it out if it fails, and plan the next experiment without repeating failures. I gave a few examples that had to pass, tight latency targets, and let it run. Sol did some amazing things. First, it scanned benchmarks and shortlisted base models: Qwen 3.5, Gemma 4, Liquid LFM 2.5. It found a dataset on HuggingFace for typed text. Then it built a simulator for fingers striking a Mac keyboard, modeling the physical layout with a Gaussian distribution around each key. It simulated striking the wrong key, wrong order, fat-fingering, etc. With the models + data + simulator, it fine-tuned using MLX right on my MacBook. It had a working prototype within an hour! But accuracy was pretty poor. — Problem 1: Tokenization Sol read papers, ran tests, and identified that the tokenizer was the bottleneck. Tokenization makes typos hard for the model to see, so it memorizes mappings instead of using its language priors. Sol tried ByT5, Google’s tokenizer-free byte-level LLM. This made a big improvement, but the model is old and lacked the knowledge needed to reach Sol performance. Sol dug deeper and realized a tokenizer-free model isn’t needed; instead, it used T5Gemma, an encoder-decoder model. This can understand the input deeply before producing output, and furthermore, Sol could post-train the encoder to improve performance. This gave a much higher ceiling. — Problem 2: Loss function Now the model was correcting some typos perfectly, but ignoring most. Sol realized that standard cross-entropy loss was teaching the model to avoid edits, because the vast majority of characters in the training data were left unmodified. The fix was wild: Sol wrote a custom loss function that byte-aligns the source and target strings, uses a dynamic programming algorithm to compute the minimum edits between the two, then weights correct edits much higher than copies. After a lot of tuning, this dramatically improved accuracy. — Problem 3: Autoregression One failure mode remained: if the model made a mistake, it couldn’t backtrack. It could only predict the next token. Teaching it to “think” like a reasoning model would solve this, but would be far too slow. Sol found a beautiful solution: instead of greedily predicting the next token, beam search over all possibilities. This parallelizes the exploration instead of one linear chain-of-thought. At the end, choose the path with highest cumulative log probability. This worked great, but made the experience worse, since the user wouldn’t see progress until the whole search was done. To fix this, Sol made a clever observation: after each search step, the longest common prefix among surviving branches is guaranteed to appear in the final result, so it can be displayed immediately. As the search progresses, weaker paths are dropped and the prefix grows, so the user sees continuous progress. Sol built all this as a custom MLX pipeline that does the parallel decoding on the MacBook GPU, with just ~40ms TTFT. It’s crazy fast and entirely local. — Final eval (error reduction rate, higher is better): - Apple autocorrect: 49.66% - GPT-5.6 Luna: 82.47% - GPT-5.6 Terra: 87.64% - GPT-5.6 Sol: 90.56% - Our model (1.7B): 91.02% Final cost: - 1 quota reset (thanks Tibo) - $0 (And yes, I verified there's no cheating. In fact, we test words scrubbed from the training data to prove the model isn’t memorizing) There were a ton more details and tangents I could write about: contrastive learning, GRPO, DPO, dynamic masking, and more. Sol is a fascinating and creative model. It blew my mind so many times. Don’t let a lack of experience stop you: Sol makes AI experiments accessible to anyone!

This is my "feel the AGI" moment: I used GPT-5.6 Sol to train my own autocorrect model that outperforms GPT-5.6 Sol (wtf??) I have no ML background. I have no idea what I'm doing. I just kept pushing Sol until it spat out a SOTA model. And I spent $0. The motivation: Years of talking to AI have made me terrible at typing. Rather than fix my skill issue, I decided to throw more AI at it. My idea was: instead of autocorrect that interrupts my flow, I want to type fast with mistakes and have AI clean it up after. I wanted the smallest local model possible, for speed, for battery life, for science! So I decided to train my own. Inspired by Andrej Karpathy’s autoresearch, I ran Codex /goal with this setup: pick an experiment, try it, record the results to a doc, throw it out if it fails, and plan the next experiment without repeating failures. I gave a few examples that had to pass, tight latency targets, and let it run. Sol did some amazing things. First, it scanned benchmarks and shortlisted base models: Qwen 3.5, Gemma 4, Liquid LFM 2.5. It found a dataset on HuggingFace for typed text. Then it built a simulator for fingers striking a Mac keyboard, modeling the physical layout with a Gaussian distribution around each key. It simulated striking the wrong key, wrong order, fat-fingering, etc. With the models + data + simulator, it fine-tuned using MLX right on my MacBook. It had a working prototype within an hour! But accuracy was pretty poor. — Problem 1: Tokenization Sol read papers, ran tests, and identified that the tokenizer was the bottleneck. Tokenization makes typos hard for the model to see, so it memorizes mappings instead of using its language priors. Sol tried ByT5, Google’s tokenizer-free byte-level LLM. This made a big improvement, but the model is old and lacked the knowledge needed to reach Sol performance. Sol dug deeper and realized a tokenizer-free model isn’t needed; instead, it used T5Gemma, an encoder-decoder model. This can understand the input deeply before producing output, and furthermore, Sol could post-train the encoder to improve performance. This gave a much higher ceiling. — Problem 2: Loss function Now the model was correcting some typos perfectly, but ignoring most. Sol realized that standard cross-entropy loss was teaching the model to avoid edits, because the vast majority of characters in the training data were left unmodified. The fix was wild: Sol wrote a custom loss function that byte-aligns the source and target strings, uses a dynamic programming algorithm to compute the minimum edits between the two, then weights correct edits much higher than copies. After a lot of tuning, this dramatically improved accuracy. — Problem 3: Autoregression One failure mode remained: if the model made a mistake, it couldn’t backtrack. It could only predict the next token. Teaching it to “think” like a reasoning model would solve this, but would be far too slow. Sol found a beautiful solution: instead of greedily predicting the next token, beam search over all possibilities. This parallelizes the exploration instead of one linear chain-of-thought. At the end, choose the path with highest cumulative log probability. This worked great, but made the experience worse, since the user wouldn’t see progress until the whole search was done. To fix this, Sol made a clever observation: after each search step, the longest common prefix among surviving branches is guaranteed to appear in the final result, so it can be displayed immediately. As the search progresses, weaker paths are dropped and the prefix grows, so the user sees continuous progress. Sol built all this as a custom MLX pipeline that does the parallel decoding on the MacBook GPU, with just ~40ms TTFT. It’s crazy fast and entirely local. — Final eval (error reduction rate, higher is better): - Apple autocorrect: 49.66% - GPT-5.6 Luna: 82.47% - GPT-5.6 Terra: 87.64% - GPT-5.6 Sol: 90.56% - Our model (1.7B): 91.02% Final cost: - 1 quota reset (thanks Tibo) - $0 (And yes, I verified there's no cheating. In fact, we test words scrubbed from the training data to prove the model isn’t memorizing) There were a ton more details and tangents I could write about: contrastive learning, GRPO, DPO, dynamic masking, and more. Sol is a fascinating and creative model. It blew my mind so many times. Don’t let a lack of experience stop you: Sol makes AI experiments accessible to anyone!

Anshu

177,702 次观看 • 11 天前