AI Briefing
TLDR AI news
The latest AI and developer news from TLDR AI, with short summaries.
Visit website →OpenAI Adds Animated Pets and Config File Import to Codex
OpenAI has added animated Pets and a config file import feature to Codex.
Google Testing New Omni Video Generation Model
Signs have emerged that Google is testing a new Omni model in the Gemini video generation tab.
Anthropic tests Jupiter-v1-p ahead of developer conference
Anthropic appears to have started red teaming Jupiter V1 ahead of its May 6 developer conference.
Advanced Quantization Algorithm for LLMs
AutoRound has strengthened 2-4 bit LLM/VLM quantization and ecosystem integration.
Speculative Decoding for RL Training
Applying Speculative Decoding to NeMo-RL boosted RL rollouts by 1.8x.
What actually goes into a SKILL.md
SKILL.md is not a prompt but a loader specification, and structure determines context cost.
Silico (3 min read)
Goodfire has released an early preview of Silico, an interpretability-based AI model design platform.
Continuous Improvement of the Agent Execution System
Cursor improved agent quality by customizing the execution system per model and using dynamic context.
AI Boom Turns Memory Chips Into One of the World's Most Profitable Products
The AI boom has triggered a super cycle in the memory chip industry, driving Samsung Electronics to record-breaking earnings.
Perplexity Expands Enterprise AI Workflows
Perplexity has added Teams, Excel, Workflows, and new data connectors to Computer for Enterprise.
Why LLM Serving Needs to Separate CPU and GPU
SMG separated serving CPU work into a Rust gateway to eliminate the Python GIL bottleneck.
AWS Neuron SDK unveils Neuron Agentic Development for NKI kernel development for Trainium
AWS Neuron has unveiled Neuron Agentic Development for developing NKI kernels for Trainium.
GLM-5V-Turbo: Toward a Foundation Model for Multimodal Agents
GLM-5V-Turbo is an agent model that integrates multimodal perception, tool use, and end-to-end verification.
New frontier AI models got faster at spatial biology, but accuracy stayed flat
On SpatialBench, GPT-5.5 and Opus 4.7 got faster but accuracy barely moved.
Cursor's War Chest, xAI's Twist
Cursor's sale to xAI showed that the independence of AI coding apps has hit its limit.
KV Cache Locality: The Hidden Variable in LLM Serving Costs
Per-GPU KV cache locality determined LLM serving cost and TTFT.
Anthropic nears $900 billion valuation round
Anthropic is closing in on raising $50 billion at a $900 billion valuation.
xAI Unveils Grok 4.3
xAI unveiled Grok 4.3, boosting both benchmark scores and cost efficiency.
Claude Security Enters Public Beta
Claude Security has moved into public beta for Claude Enterprise.
Where Did the Goblins Come From
OpenAI traced the cause of the goblin-expression habit that spread after GPT-5.1 to the Nerdy personality's reward mechanism.