Which kind of AGI parent are you?
AI
-
Project Glasswing: Limited Distribution AI Initiative Unveiled
By
–
correct: not publicly released, but see Project Glasswing for limited distribution
-
AI hasn’t produced a single paperclip yet in 2026
By
–
it's 2026 and ai has not even made a single paperclip
-
PDR Framework: Parallel Reasoning Agents for Complex Scientific Queries
By
–
Reasoning doesn’t have to mean longer chains of thought:
— Anirudh Goyal (@anirudhg9119) 8 avril 2026
PDR = draft in parallel → distill into a compact workspace → refine, and shift the Pareto frontier.https://t.co/4Sca6dFu4Q https://t.co/kk0fYpc8Y1 pic.twitter.com/PvevaX9ngYReasoning doesn’t have to mean longer chains of thought: PDR = draft in parallel → distill into a compact workspace → refine, and shift the Pareto frontier. arxiv.org/abs/2510.01123 Alexandr Wang (@alexandr_wang) 3/ we’re also releasing contemplating mode, which orchestrates multiple agents that reason in parallel designed to handle complex scientific & reasoning queries. in our testing we found it competitive w/ other extreme reasoning models such as Gemini Deep Think & GPT Pro. — https://nitter.net/alexandr_wang/status/2041909381667958855#m
-
Muse Spark: Multi-Agent Collaboration for Test-Time Reasoning Scaling
By
–
To spend more test-time reasoning without drastically increasing latency, we can scale the number of parallel agents that collaborate to solve hard problems. While standard test-time scaling has a single agent think for longer, scaling Muse Spark with multi-agent thinking enables
-
Muse Spark enables predictable scaling toward personal superintelligence
By
–
With Muse Spark, we are on a predictable and efficient scaling trajectory. We look forward to sharing increasingly capable models on the path to personal superintelligence soon.
-
Reinforcement Learning Stack Achieves Stable, Predictable Model Capability Gains
By
–
Reinforcement learning leverages compute to scalably amplify model capabilities. Though large-scale implementation is often prone to instability, our new stack delivers smooth, predictable gains, showing log-linear growth in pass@1 and pass@16 (at least 1 success across 16
-
Reinforcement Learning Optimizes Model Reasoning with Token Efficiency
By
–
RL trains our models to "think" before they answer, a process known as test-time reasoning. To serve this capability to billions of users and efficiently use tokens, we rely on two key levers: thinking time penalties to optimize token use and multi-agent orchestration that boosts… pic.twitter.com/oHHap4NAg3
— AI at Meta (@AIatMeta) 8 avril 2026RL trains our models to "think" before they answer, a process known as test-time reasoning. To serve this capability to billions of users and efficiently use tokens, we rely on two key levers: thinking time penalties to optimize token use and multi-agent orchestration that boosts
-
Scaling Properties of Muse Spark: Pretraining, RL, and Reasoning
By
–
To build personal superintelligence, our model’s capabilities should scale predictably and efficiently. Below, we share how we study and track Muse Spark’s scaling properties along three axes: pretraining, reinforcement learning, and test-time reasoning. Let’s start with
-
Meta’s Muse Spark Converts Images to Code with Asset Extraction
By
–
Ok this is actually pretty impressive and I truly didn't see any model doing this before or being able to do it to this extent.
— Pietro Schirano (@skirano) 8 avril 2026
When I asked Muse Spark from Meta to convert this image into code, it cut out the assets from the screens so it could use them correctly! pic.twitter.com/eyTlSHk2BhOk this is actually pretty impressive and I truly didn't see any model doing this before or being able to do it to this extent. When I asked Muse Spark from Meta to convert this image into code, it cut out the assets from the screens so it could use them correctly!
→ View original post on X — @alexandr_wang, 2026-04-08 17:08 UTC