全部 AI 动态

The Batu Khan was made with Seedance 2.5 on PixVerse 30s winter standoff: Batu’s cavalry. A Rus shield wall. Snow. Breath. RT + reply “BATU” → full workflow in DMs

Muse Glimmer one-shot 5 games! @AIatMeta dropped Muse Glimmer. So we ran it against Muse Spark 1.2, on five one-file game demos, each one written to play itself: Tetris, a top-down pixel street race, a rooftop web-slinger, a blue hedgehog platformer, and Flappy Bird. the setup: • Muse Glimmer 30B — via aimlapi[.]com. cost: $0.02 • Muse Spark 1.2 — via aimlapi[.]com. cost: $0.11 what we saw: • every Glimmer scene runs and plays itself. it's a reasoning model, so it burns more tokens — but the output is clean. its Tetris even scores each placement to keep lines clearing, and the blocks burst into pixels on every clear. • Glimmer beat Spark at Flappy Bird — its bird flew through the pipes cleaner. • but it's not perfect: in the racing game its cars sometimes slip off the road, and its graphics are simpler and blockier than Spark's. glad to see Meta back in open source — and we're waiting for Spark 1.2 weights next 🙌

🚀 MiniMax H3, accelerated on Day 1 with Sol Engine! Within just 4.5 hours, our agent-native Sol Video Inference Engine achieved: ⚡ 3.95× end-to-end speedup over Diffusers ⚡ 2.80× speedup over SGLang 🎬 8× NVIDIA GB200, 1344×768, 24 FPS, 124 frames The acceleration combines kernel fusion and graph capture, cross-step caching, and training-free sparse attention powered by Sol-Attn—with no distillation, fine-tuning, LoRA, or offline calibration. We are especially excited to see a powerful open-weight model like MiniMax H3 released to the community. Open models are essential for pushing video generation research, systems optimization, and real-world deployment forward. We hope this is the beginning of a much more vibrant open-source video generation ecosystem—and Sol Engine will keep working to make the latest models faster and easier to deploy from day one. 🔗https://nvlabs.github.io/Sana/Sol-Engine/H3/

ive been playing with @AntLingAGI's ling 3.0 flash 124b for a week now, and i need to publish some numbers and performance i found valuable from this lab. a week ago the weights dropped. since then, on one dgx spark, i benched the community quant ladder, ran every official quant to ground, took the official int4 from doesn't run to the fastest single dgx spark path, watched the model write working software in one shot through an agent loop, and served its full 256K context. everything measured the same way and public, nothing tuned to flatter anyone. the verdict: ling 3.0 flash earned a permanent seat on my box. it's the fastest 124B class path i've measured on a single spark, 38.7 tok/s on the official int4 recipe, 35.2 on the community gguf, 2.4x what deepseek v4 flash does on the same hardware, and it writes correct code on the first pass often enough to feel like a different class of local model. dive with me into the full week of receipts, quant by quant, curve by curve 👇

New Muse Glimmer 30B destroyed Gemma 4 31B at making retro arcade games! We gave three models the same task and compared one-shot outputs. Muse Glimmer 30B, Gemma 4 31B and Qwen3.6 27B each ran locally on its own RTX 5090 with 32GB VRAM Tasks: - Space Invaders - Tetris - Arkanoid with multiball Outputs: Muse Glimmer 30B: 83.4K tokens, 17.7 min Qwen3.6 27B: 23.7K tokens, 5.4 min Gemma 4 31B: 17.8K tokens, 4.5 min Glimmer's games play like the originals. The alien fleet speeds up with every kill, the blocks lock and lines clear, the bricks break. Nothing froze or crashed. Gemma broke on Tetris, its bot piles every piece against the left wall until the stack hits the ceiling with half the well empty. Qwen finished all three but some of its Arkanoid balls phase straight through the paddle. Run Muse Glimmer yourself in Atomic Chat from day 0!

Testing Grok Imagine 2.0 against other image models: Prompt: "A mathematician drinking from an object of genus 1" ChatGPT gets trapped by the donut analogy: it turns a donut (genus 1) into a mug, accidentally creating an object of genus 2. Grok gets it right: a regular coffee mug already has genus 1.

Fast H3 implementation for Metal. Enjoy, modify, and so forth: https://github.com/antirez/h3.c Contains code from @liuliu which is welcomed in taking back whatever parts he likes for @drawthingsapp in case there are H3 plans there.

Imagining a space cruise with Flux 3: > amateur handheld pov footage of a tourist in their plush and comfortable room of a space cruise, carpeted floor and comfy bed, large window shows a view of Saturn, you can see their reflection in the window, they turn off the light half way through so we can see outside better

Learn 99% of ChatGPT Work in 61 minutes: GPT Work is like Codex in the cloud. It works on phone, web, and desktop. And I've been using it to run my business. These are the 14 knowledge work capabilities. 00:00 Intro 03:18 #1 Presentations 09:16 #2 Plugins 17:36 #3 Blocks 23:34 #4 Websites & Apps & Hosting 30:09 #6 Branching Chats 31:51 #7 Desktop App (Cloud vs Local) 37:01 #8 Voice Mode 41:36 #9 Remote Voice Mode 45:52 #10 In-App Browser 48:50 #11 Skills 51:45 #12 Scheduled Automations 54:37 #13 Spreadsheets 56:58 #14 Multi-Agent Workspace 59:19 Final Review

Used ChatGPT finance just now and asked it to find phantom subscriptions. It unearthed $550 a year of random things I didn’t realize I was paying for and thought I canceled. Thanks @OpenAI and whoever is on the Finance product.

We re-tested GPT-5.6 Luna from @OpenAI on ARC-AGI (Verified) following its recent 80% price reduction: - ARC-AGI-2: 59.6%, $0.18/task - ARC-AGI-1: 90.7%, $0.07/task The new results match Luna's original performance at a much lower cost.

Her first message to me in slack was: @cursor_ai /pstack investigate and fix

It's finally here! Minimax H3 Turbo lora makes generations 5x faster. Use only 4 steps instead of 20. https://huggingface.co/larryvrh/MiniMax-H3-Turbo-Lora Also see: https://huggingface.co/larryvrh/MiniMax-H3-Turbo-Lora/discussions/6

Muse Spark 1.2 places Meta on the Cost per Task Pareto frontier, scoring 6 points below Claude Opus 5 at ~1/6th of the cost At Meta's $1.25/$4.25 per 1M token pricing, Muse Spark 1.2 (xhigh) sits on the Pareto frontier of Intelligence Index vs Cost per Task. It delivers comparable intelligence to Claude Opus 4.8 (max, $2.03) at a fifth of the cost per task, and undercuts GPT-5.6 Sol (high, $0.55), GPT-5.6 Terra (max, $0.61), and Kimi K3 (max, $0.87). The nearest cheaper options are Grok 4.5 (high, $0.36) and GPT-5.6 Sol (medium, $0.37), and they all sit below it on the Index. The step up from Muse Spark 1.1 ($0.29 per task) comes at unchanged per-token pricing, with the increase driven by heavier token usage on agentic tasks.

Grok 4.5 just beat Kimi K3 at 13x cheaper price! $0.15 vs $1.98 — same task, same prompt, one shot. We'll miss Nikita Bier — the best Head of Product X ever had. So we had four models honor him the only way AI can: by building a 3D plush Nikita Boar in his signature suit 🐗 Cost per figure: Grok 4.5 — $0.15 GPT 5.6 Sol — $0.60 Qwen 3.8 Max — $0.67 Kimi K3 — $1.98 Kimi spent ~19 minutes thinking and billed 13x more. Grok just shipped. Everyone argues about which model is smartest. Nobody checks the receipt. And then we made the hogs dance with Grok 🕺 All four models, one API key — AI/ML API, 1000+ models.