Exciting news: Muse Spark Open (Max) by @AIatMeta is the #1 US open model in the Code Arena: WebDev! With 1511 pts, it’s #13 among open models overall. As the first open-weight release in the Muse Spark family, it scored 156 pts above Muse Glimmer, @AIatMeta ’s earlier open model at #41 among open models with 1355 pts. Priced at $1.25/$4.25 per 1M input/output tokens, it’s 79% less than Muse Wonder (Max) and priced on par with Muse Spark 1.3 (Max). By domain, Muse Spark Open (Max) ranks: - #13 in Brand & Marketing, and in Data & Analytics - #14 in Reference-Based Design - #15 Gaming, and in Consumer Product - #20 in Simulations Congrats to the team at @AIatMeta on this contribution to the open ecosystem! AI at Meta @AIatMeta Today we’re releasing Muse Spark Open, the most capable US open-weight model to date. Built for coding and agentic work, it excels at long-horizon tasks, complex software engineering, and multi-step reasoning. Key Model Specs: → 816B total / 42B active parameters → 1M token context window → Natively multimodal across text, image, video, audio 🔗 View Quoted Tweet
ARCHIVE
内容归档
正文、图片和 Markdown 均已本地保存，不包含回复区。
We’re introducing Bring Your Own Cloud (BYOC) for Replit, with options for development and production in your cloud, whether you use @Azure , @GoogleCloud , or @AWScloud . Our approach gives security and platform teams greater isolation and control while preserving the familiar Replit experience. Explore BYOC: replit.com/blog/bring-you…
x.com/ModelScope2022… ModelScope @ModelScope2022 NVIDIA Nemotron 3.5 Super VL is open released in BF16 and NVFP4: a 120B MoE multimodal agent model that activates just 12B parameters per token. ModelScope’s ms-swift supports both inference and fine-tuning, helping developers put the model to work and adapt it to specialized multimodal applications. 🤖 modelscope.ai/collections/nv… 8 � github.com/modelscope/ms-… Ms 🏆 Agent performance: Outperforms the 5x larger Nemotron 3 Ultra on the reported Hermes Bench and LangChain evaluations. ⚡ Faster workflows: Delivers up to 5x higher throughput in its size class and completes evaluated agent tasks up to 4x faster. 👁️ Multimodal reasoning: Understands text, images, charts, documents, and video while reasoning, coding, and using tools. 🧰 Built to customize: Fine-tune with your own multimodal datasets using LoRA in ms-swift, or use NVIDIA NeMo for SFT and RL to adapt the model to domain data, tools, policies, and workflows. 🧠 Efficient deployment: NVFP4 with QAD, multi-token prediction, and cross-turn context reuse reduce repeated inference work. Supports vLLM, SGLang, TRT-LLM, and NIM across local, data center, and cloud environments. 🔗 View Quoted Tweet
Harness durability in fx.sh . The same 𝚌𝚛𝚎𝚊𝚝𝚎𝙵𝚡𝙰𝚐𝚎𝚗𝚝() is now *durable* on your laptop or in the cloud, via files, Postgres, or Vercel Workflows. Node.js or Bun. Agents need long-running compute. But they also need to tolerate restarts, redeploys, timeouts, crashes… Concretely, if a user sends a prompt and then throws their phone into a black hole, we ensure that turn runs to completion. If the process crashes mid-tool-call, it resumes from there. To this end, 𝚕𝚒𝚋𝚏𝚡 introduces a simple durability adapter API. On Vercel, it leverages all the infra we built for durable workflows and queues, giving you reliability with 1LOC. This is a critical aspect of production-grade agents (and something all @eve agents also get out of the box). Next, we'll show you how we bring the same durability to Pi, OpenCode, and even your own custom harnesses. Pranit @fazxes fx v0.0.14 libfx is now durable out of the box. Same API on local, Vercel and DIY. - Sessions survive crashes, redeploys & timeouts - Power cloud agents with ChatGPT & Grok subscriptions - Custom providers & local model support in libfx - Run multiple libfx agents in one app - More control over each libfx session - TypeScript types for libfx's public APIs - Ported to Zig 0.17 - Multiple bug fixes & security improvements fx.sh/changelog#v0.0… 🔗 View on Twitter 🔗 View Quoted Tweet
Why did code quality matter in the first place? @mkanat brings it back to two questions: what does the system actually do, and how do you change it safely? 57 seconds from AI Engineer New York. Before you accept the next AI generated change, can you answer both? #AIEngineerNY 🔗 View on Twitter
Here’s how our engineers use Codex at OpenAI to dig through our codebase, find what needs attention, and chip away at the maintenance backlog: Rodrigo Roim @rrroim Software maintenance is changing. A few months ago, I had to tell Codex exactly where to look for dead code. Now it can find candidates in Slack, check them against production telemetry, and draft cleanup PRs. Here’s how we do this at OpenAI. developers.openai.com/blog/maintaini… 🔗 View Quoted Tweet
Congrats @alexandr_wang and @Muse team on releasing such an epic open weights model! Artificial Analysis @ArtificialAnlys Meta has released Muse Spark Open, an 816B (42B active) open-weights model that scores 39 on the Artificial Analysis Intelligence Index, making it the #1 US open-weights model Muse Spark Open (xhigh) scores 22 points above Muse Glimmer (high, 17), Meta's 30B dense open-weights model from August. It is also ahead of NVIDIA's Nemotron 3.5 Super VL (26), the next highest-scoring US open-weights model, which launched today @AIatMeta shared access with us ahead of public release for benchmarking. Congratulations to @AIatMeta , @finkd , and @alexandr_wang on the release! Key Takeaways: ➤ Open weights at the level of Meta's August flagship. Muse Spark Open scores within a point of Muse Spark 1.2 (xhigh, 40). It is stronger on AutomationBench-AA and has a lower hallucination rate, but weaker on GDPval-AA v2.1, CritPt and AA-Omniscience accuracy. Within Meta's current lineup, it sits 13 points behind Muse Wonder (max, 52) and 17 points behind Muse Wonder Full (max, 56) ➤ Token-efficient, at a moderate cost per task. Muse Spark Open uses 40k output tokens per Intelligence Index task, the fewest among the nine highest-scoring open-weights models: e.g. 16% fewer than Muse Spark 1.2 (xhigh, 48k) and 43% fewer than GLM-5.3 (max, 71k), its closest Chinese peer by size. Together AI, Fireworks and Baseten offer it at launch for $1.25/$4.25 per 1M input/output tokens ($0.15 cached input). At that pricing it costs $0.84 per Intelligence Index task. That is less than Muse Spark 1.2 ($0.97) at the same per-token prices and less than half of GLM-5.3 (max, $2.01), but about three times DeepSeek V4.1 Flash (max, $0.27) ➤ A relative weakness in agentic work compared to Chinese open-weights leaders. Among the nine highest-scoring open-weights models, Muse Spark Open ranks last on the four agentic evaluations in the index but fourth on the other six combined, and it leads all open-weights models on AA-Omniscience (25). GLM-5.3 (max), the closest Chinese model by size, owes all of its 6-point lead to the agentic evaluations ➤ Self-hostable on a single node, or two Mac Studios. BF16 weights are ~1.6TB. Meta's NVFP4 checkpoint quantizes only the 256 experts, cutting the weights to ~530GB. Meta says this runs on four Blackwell Ultra GPUs or eight H200s, or on two 512GB M5 Ultra Mac Studios using MLX. The architecture uses 62 layers with grouped-query attention (16 KV heads), 8 of 256 experts active per token, and a 1.8B-parameter Perception Encoder for vision Other model details: ➤ Parameters: 816B total, 42B active (MoE) ➤ Context window: 1M tokens ➤ License: Muse Model License, a custom license that requires "Built with Muse Spark Open" attribution and a separate agreement with Meta for Model-as-a-Service businesses with over $20M in annual revenue ➤ Input modalities: text, image, video and audio. Output: text ➤ Availability: weights on Hugging Face (BF16 and NVFP4), with API access via Baseten, Fireworks AI, Groq, Modal, Nebius, OpenRouter and Together AI 🔗 View Quoted Tweet
这效率 Roadster 游戏都出来了 roadster-rush.ai.studio Ammaar Reshi @ammaar I used Gemini 4 Argon to make Subway Surfers for the new flying Tesla Roadster 🚀 Fly over moving traffic, hit the thrusters to barrel roll in the sky, and collect air power-ups to set a high score. Play + code below! 🔗 View on Twitter 🔗 View Quoted Tweet
Openrouter is the best place to use Perplexity Decider models OpenRouter @OpenRouter Perplexity Decider v1.2 and Decider Flash v1.2 are now on OpenRouter @perplexity_ai 's open-weights decision models take text or images and return typed answers with probabilities - Decider v1.2: a new 27B checkpoint, succeeding v1.1 - Decider Flash v1.2: 4B, faster and half the cost openrouter.ai/perplexity/ppl… openrouter.ai/perplexity/ppl… 🔗 View Quoted Tweet
Fascinating to see Mathematicians go through all the stages of grief — Denial, Anger, Bargaining, Depression, and Acceptance — in such a compressed window of time. We're at the tail end of depression and entering acceptance. Haider. @haider1 Mathematician Terence Tao: "we're seeing the end of the romantic era of mathematics, the age of heroes. we're moving into an industrial era where math becomes large-scale, done in big teams like the other sciences" We're entering, rather unwillingly, a new era where exciting new things are possible, but some things are being lost 🔗 View on Twitter 🔗 View Quoted Tweet
In @Nature this week, our new research shows how machine learning can improve weather forecasts two to six weeks out, a timeframe that's historically been very difficult to predict. It's exciting to see the potential of this work to help us better prepare for extreme weather, manage natural resources, and make more informed decisions in agriculture. There's a broader lesson here... the future may not just be about ever-more-capable models, but about building intelligent verification systems around them: systems that learn where models are systematically wrong, reason about cause and effect, and continuously calibrate predictions against reality. Read more: microsoft.com/en-us/research…
In Aisle's first year, they've shown that the system built around an LLM is highly important, and even small, cheap fast models can be boosted to do valuable security work Stanislav Fort @stanislavfort My AI security research lab AISLE discovered 500+ zero-days with CVEs across 100+ codebases in a year => ~1% of 2026's open-source CVEs globally from a single team! On hardened codebases like curl, we've repeatedly matched or outperformed frontier AI labs in zero-day discovery 🔗 View Quoted Tweet
Demis Hassabis on choosing what’s worth spending your life building in a world changing this fast: “Going after hard problems and deep problems, it’s no more difficult in some ways than going after a shallower, simpler, more superficial problem. They’re just differently difficult.” “Given life’s very short and you only have so much time and energy, you might as well put your life force into something that will really make a difference.” “If you start off on a deep tech journey today... you’re talking about a 10-year journey for true deep tech. So now you have to consider AGI appearing in the middle of that journey.” “Imagine what that world would look like and then build something that would be useful if that comes in halfway through.” 🔗 View on Twitter
This Week in Replit 1) Introducing Runway for web design. Connect Runway to Replit to generate images, video, and assets directly inside your project, from first idea to finished site. Turn a product photo into a commercial, make UGC-style ads and launch videos, or create image and video assets for your website. → runway.com/mcp#replit 2) Introducing the Replit Solutions Partner Program, for the systems integrators, agencies, and consultancies who turn client briefs into working software on Replit. It brings training and a referral-first partnership model to the work partners are already doing. → replit.com/blog/introduci… 🔗 View on Twitter
Perplexity Decider v1.2 and Decider Flash v1.2 are now on OpenRouter @perplexity_ai 's open-weights decision models take text or images and return typed answers with probabilities - Decider v1.2: a new 27B checkpoint, succeeding v1.1 - Decider Flash v1.2: 4B, faster and half the cost openrouter.ai/perplexity/ppl… openrouter.ai/perplexity/ppl…
not the first time Dwarkesh comes as off to me as working as an undisclosed press secretary for the tech industry. Fireside Alpha @firesidealpha Dwarkesh Patel reveals an Anthropic employee told him two days ago they expect AIs to match humans at AI research within nine months based on inside information they can’t share, and says labs should make that evidence public “I think there’s a very frustrating element to the current situation right now, where people inside the AI labs claim to have evidence that makes them more worried. And they are not allowed to talk about it. And this makes it a very difficult situation on the outside to evaluate how worried we should be." "So for example, just the other day, maybe two days ago, I was talking to somebody at Anthropic and they were like, look, I expect AIs to be humans at AI research within nine months." "And I’m like, why do you believe that? Well, they’re like, it’s not just vibes, but I have information on the inside. Like, what is it? Is it a new model? Is it some evaluation you ran? It’s like, I can’t talk about it." "I think it’s a terrible state of affairs. If they really believe this, if they think that we’re in a situation where we’re less than a year away from automating AI research and radically speeding it up, they should make this evidence public.” 🔗 View on Twitter 🔗 View Quoted Tweet
"Refuse to confuse done with good." @lil_dill 🔗 View on Twitter Katie Dill @lil_dill If we only use AI to make the things that we already make, just faster, then we're definitely missing out on the most interesting part. From my talk at @lennysan ’s Summit, here are four ideas for raising the ceiling, not just the floor: - Have a point of view [5:38] - Encode your standards into the machine [9:07] - Don't confuse done with good [12:21] - Unleash creativity and artistry [16:19] Full talk here: youtu.be/GLvFTMtw4Jk?si… 🔗 View Quoted Tweet