GPT-6 Astra vs Kimi K3. I'm comparing these because... the gap between US models and Chinese models is already getting small. If US AI labs slow down, they will be the ones most affected by it. > Astra did it in one shot 6m 20s and $8. > Kimi K3 got almost the same result, but I had to use 2 follow-up prompts. Total: 12 min and $6.35. The results are almost identical. If US labs slow down, China takes the AI race. They will ship Astra-level open-weight models. China is already good at this. Open weights become the default. That's a nightmare for US labs.show more

Kimi K3 + GPT-6 Astra can become something bigger... than two agents: a Two-Brain AI Operating System the formula: Two-Brain OS = Research Brain + Execution Brain + Shared State + Router + Tools + Verification not two models doing the same job. two specialized brains connected through one persistent system step 1 -> Kimi K3 becomes the research brain. complex research, source comparison, long-context synthesis and strategic planning happen here. its job is to explore the problem and compress the findings into a clear plan. step 2 -> GPT-6 Astra becomes the execution brain. it receives the plan, writes code, operates tools, creates artifacts and turns decisions into finished work. step 3 -> build a shared state layer. store the objective, evidence, decisions, constraints, failed attempts and next action outside the chat. both brains always know what happened and where the system should continue. step 4 -> add the router. uncertainty and open-ended questions move to Kimi K3. execution, tool use and deterministic tasks move to GPT-6 Astra. every job reaches the brain designed to handle it. step 5 -> create the handoff loop: research -> plan -> execute -> inspect -> update state -> continue. if execution reveals missing information, the task returns to Kimi. if the plan is ready, Astra takes control again. step 6 -> verify before completion. tests, source checks, constraints and explicit success criteria decide whether the result ships or returns to the correct brain with a clear failure signal. that is the difference between using two AI models and building a Two-Brain AI Operating System. Kimi K3 expands the search space. GPT-6 Astra converts it into action. shared state preserves progress, the router controls every handoff, tools execute the work and verification decides when the system is actually finished. one model can generate an answer. two specialized brains connected through memory, routing and verification can run an entire workflow. the full Kimi K3 + GPT-6 Astra Two-Brain OS breakdown is below ↓show more

🇨🇳 CHINA DOESN'T NEED TO BEAT US AI. IT... JUST NEEDS TO UNDERCUT IT. The US pulled Anthropic's Fable offline, then told OpenAI to release GPT-5.6 customer by customer, government-approved. Both frontier labs now need federal sign-off to ship. China's open models can't be gated, sit a few points behind, and cost a fraction. Make it nearly as good. Give it away. Take the market. The US isn't slowing China down. It's accelerating its own downfall.show more

GPT-6 Astra vs Fable 5.1 vs Kimi K3 vs... Sol Astra: Took 11 minutes. This is from a follow-up after the first attempt had rendering and UI bugs. The smoke feels a little off. I expected more detail, but it did a good job understanding the intent. Cost: $12.85. Fable 5.1: Took 18 minutes to complete the same test with high detailing. Cost: $12.60. Kimi K3: Took 10 minutes with a similar level of output and cost only $6.95. SOL: Took 7 minutes, but the output seemed messy. Cost: $5.78. Personally, I like Kimi K3 for delivering similar quality with better token efficiency, followed by fable 5.1 for realism. What about you? I’ll do a couple more tests before coming to a conclusion.show more

a moonshot engineer leaked the benchmark anthropic, openai and... xai all buried the same week: kimi k3 beat opus 5, gpt-5.6 and grok 4.6 at $0.94 a task. stop paying anthropic $200 a month for opus 5 and openai $200 for gpt-5.6 when kimi does the same work for $8 the leak showed kimi k3 winning 9 of 12 categories against opus 5, gpt-5.6 and grok 4.6. within 48 hours all three labs quietly pushed pricing pages and one very specific comparison chart off their sites. nobody announced anything. they just deleted, which tells you everything the four numbers they scrubbed: cost per task · $0.94 vs $1.80 -> opus 5 charges $1.80 to finish one task. gpt-5.6 $1.04. grok 4.6 $0.61. kimi k3 $0.94 and it landed 487 of 500 clean -> anthropic is billing you double for a model that lost the benchmark it paid to promote the weights · free, sitting on huggingface right now -> the entire model is a public download. pull it, keep it, run it forever, nobody can switch it off -> a model you can hold cannot be rented at $200 a month. that single fact is what three labs deleted a chart over the switch · one line of bash -> moonshot ships an anthropic-compatible endpoint. one env variable and claude code points at kimi -> same cli, same keybindings, same /model. you change a url, opus 5 never knows it lost the seat the bill · $400 down to $8 -> opus 5 max plus gpt-5.6 pro is $400 a month. kimi runs the same daily work for $8 metered -> that is a 98% cut for output that beat both of them 9 categories to 3 here is the part they will fight me on: the frontier tax died the week this leaked and all three labs know it. once the weights are public the price has a ceiling, because anyone can serve the same model. anthropic, openai and xai are charging 2025 prices on a lead that ended in a benchmark they deleted instead of answered drop your $400/mo ai stack to $8. the run above is kimi k3 finishing the task opus 5 bills $1.80 for. the full breakdown is in the article belowshow more

We built high-throughput materials labs in Menlo Park to... create a loop between experiments and models. The labs generate fresh data, the models learn from it, and then help us decide what to try next. Using only 1,300 H200s, plus months of our experimental data, we mid-trained and RL’d an open-source model to surpass GPT-6 Astra on our analysis benchmark. We call it Neon. This is real footage from our lab. We’re focusing first on hard problems in materials science, including superconductors, magnets, and semiconductor materials. Read our blog posts below.show more

🇨🇳 CHINA IS ALMOST READY TO UNDERCUT THE WORLD'S... BEST AI MODELS. Six months ago, that sounded impossible. Today GLM-5.2 is open source and costs a fraction of frontier models. Zhipu's co-founder says Fable-class Chinese AI comes even sooner than people think. This is the same playbook China ran on solar, on steel, on EVs. Build it cheaper. Give it away. Take the market. If China delivers frontier AI at a fraction of the price, the economics holding up the U.S. AI market start to crack.show more

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: 👇show more

Average US AI lab: > builds the strongest model... > gets scared of its own model > spends months arguing about safety > fights other AI labs over who has the strongest model > delays the release for more safety testing > regulators ask for even more safety > switches users to a dumber model > keeps it closed source Average Chinese AI lab: > trains Kimi K3 > builds a 2.8T parameter monster > gives it a 1M context window > beats GPT-5.6 and Claude Fable 5 on some tasks > open-sources the weights > publishes the architecture > refuses to elaborate > leavesshow more

🚨🇨🇳 China's new coding model flawlessly outperforms its US... rivals Zhipu AI's new flagship AI model GLM-5.2 has become the first Chinese model to rank in the top three globally on on front-end coding abilities 🔸 Users praise it as the first open-weight model reliable enough for daily coding workflows 🔸 It runs about 48% cheaper than US alternatives 🔸 It arrives as US labs face export restrictions and users fret over surging AI costs The model shows China is neck-and-neck with the US on AI performance — and now aims to overtake such giants as Anthropic and OpenAIshow more

🤯KIMI K3 ABSOLUTELY MOGS! BEATING Opus 4.8, GPT 5.5,... and even Fable 5 in multiple benchmarks. They scored 1688 on GDPval-AA v2 🔥 This is a completely different breed of open-source models Kimi creates better games and front-end designs than Fable 5, but it's 8x cheaper! The results coming out are truly impressive! I will be testing this out further and posting multiple tests today. Stay tuned!show more

Nearly a year of Code Arena: WebDev progress compressed... into 15 seconds. Each line follows the highest-scoring model from top labs over time, showing the pace of improvement across the ecosystem. In just the last year, the model in the leading spot increased score by +340 pts and the number of frontier labs competing for the top spot expanded from 6 to 10. Anthropic has dominated throughout the year. Although standout releases have jumped to the top spot, most notably the Chinese open-source model Kimi K3 from Kimi.ai in July. Today, GPT-6 Astra by OpenAI leads with 1796 pts, followed by Claude Fable 5.1 at 1764 pts. The next closest lab is 103 pts away, Qwen with 1685 pts. Code Arena: WebDev ranks models through head-to-head user preference on real front-end web development tasks. These votes drive the leaderboard that is tracking the frontier.show more

if you use Codex and you're stuck on GPT... models only, this fixes that. it's called codex-router, open source. drops other models straight into your normal Codex picker, right next to the GPT ones you already have. what it adds: - Grok, Kimi, Deepseek, Claude, all in the same picker - oauth login per provider, no api key needed - your GPT models and Chatgpt login stay untouched setup: point Codex at the repo, let it read the readme, it installs itself one snag: if you've got the Chatgpt app and a separate Codex on your path, you can end up running two different Codex versions, and the older one can choke on newer config it doesn't recognize. if the install looks off after, that's probably why, reinstall clean.show more

Introducing freeLLM v1.0 - the FREE AI models directory... People still pay $20-200/mo while FREE frontier APIs sit right next to them I collected 900+ ways to get FREE access: > 900 models across 70 providers > GPT-6 Astra, Claude Fable 5.1, DeepSeek V4, Kimi K3 for FREE > Daily updates as new promos drop > Old dead ones filtered out automatically > 20K users hit it in the first month I do the boring part (filtering noise, tracking promos, killing dead links) so you don't have to if you have suggestions for the project or spot a promo I missed, drop it in the comments and I'll add itshow more

At Open AGI, our CEO Ahmed Rashad shared why... trust is the new compute. AI is moving from generating content to making life-or-death decisions — medical diagnoses, autonomous vehicles, military systems. If the training data is wrong, the decisions are wrong. Meanwhile, models are increasingly trained on synthetic outputs, creating feedback loops that degrade quality. Data supply chains are opaque. Adversarial attacks are real. The next AI race won’t be won by bigger models. It will be won by those who control and verify the data layer. Perle Labs is building the sovereign, human-verified, on-chain auditable infrastructure high-stakes AI requires. Trust wins the AI race. Thank you Sentient for having us! No better way to end off our time at ETHDenver 🏔🦬🦄. Until next time 🫡show more

We CANNOT WAIT for you to experience Astra, the... first-ever AI creative upscaler for video. Thank you to everyone who has liked, shared, and commented—the response to Tuesday’s announcement has blown us away. If you haven’t heard back from us yet on your early access invite, you can enter your email here to make sure you’re on the list: (We’ve got thousands of comments to reply to, which is an amazing thing, but it does mean our responses will be a bit delayed. Thanks for your patience.) Astra is now in development. You’ve probably seen a few creators posting before-and-afters. We’re partnering with this group to test Astra and get valuable feedback. (Thanks, team!) Once it’s time to share Astra with early access users, you will be the first to know. Keep an eye on your email and make sure you’re following our account here for updates. - The Topaz Labs teamshow more

Many people have asked how I make my original... videos look so clear, smooth, and polished. For video upscaling and enhancement, I consistently rely on the suite of tools available from Beth Allison Labs. The two models I use most often are Astra 2 and Starlight Precise 2.5. Rather than competing with each other, they serve different purposes. There’s no single “best” model, just the right tool for the specific job. If you're creating content with Seedance 2.0, these upscaling models make an excellent addition to your workflow. Below, you'll find examples processed with Astra 2, and in the comments I'll share results from Starlight Precise 2.5. Astra 2 is Topaz's next-generation creative video upscaling model, featuring adjustable enhancement levels and prompting capabilities. It excels at correcting common AI-generated video issues such as warped details, flickering, and inconsistent textures, helping AI content appear more polished and realistic. Best used for: • AI-generated videos that need significant detail enhancement • Footage with artifacts, distortions, or missing textures • Projects that benefit from additional creative reconstructionshow more

🇺🇸 🇨🇳 CHINA JUST TOLD THE US TO KEEP... THEIR OLD NVIDIA CHIPS. Because they're smuggling the newest ones in. 🇻🇳 Taiwanese prosecutors just arrested three people for smuggling $NVDA AI chips into China. They forged documents to ship 50 Super Micro servers loaded with advanced Nvidia silicon to China, Macau, and Hong Kong. Some already cleared customs. Same company (Super Micro) whose US employees were indicted in March for allegedly diverting BILLIONS in Nvidia chips to China. The black market for Nvidia silicon is now one of the most lucrative trades on earth.show more

Some folks are saying Deep Seek censors certain topics... (like Tiannamen Square) because it's a Chinese app. This is not true. Deep Seek is an open source LLM. You can download it, or host it outside China, and use it however you want. If you use its chat interface that's based in China, it will have to comply with the Chinese regulations. But outside China, you can do pretty much anything. For our app Research Kick, we are hosting it in the US. You can see, it answers my question about Tiannamen Square.show more

Moonshot AI is casually giving developers free daily access... to Kimi K3 😳 no subscription no upfront payment just sign in and start using one of the largest open AI models available what you get for $0: - Kimi K3 with 2.8T parameters - 1M token context window - strong coding and reasoning performance - native vision capabilities - free daily credits that refresh automatically why this is worth checking: > access a frontier model without paying API fees > long context for large codebases and documents > works on web, desktop, mobile, and CLI getting started takes less than 2 minutes: 1. go to 2. create a free account 3. Kimi K3 is available as the default model 4. start chatting or coding with your daily free credits bonus: Moonshot Together lets you invite friends for a chance to earn 3, 7, 15, 30, or even 365 days of Kimi Membership through its rewards program benchmark highlights: > 2.8T parameter MoE model > 1M context window > strong performance across coding, browsing, and reasoning benchmarks important: free credits reset daily, rate limits apply on the free tier, and the open-weight release is expected on July 27 A simple way to try one of the latest frontier AI models without paying for API accessshow more