
Google Opens Lyria 3.5 to Developers, Generating Full Songs With Vocals
Google's newest music generation model expands from Flow Music into the Gemini API, AI Studio, and the Gemini app with 44.1 kHz stereo output.
20 stories

Google's newest music generation model expands from Flow Music into the Gemini API, AI Studio, and the Gemini app with 44.1 kHz stereo output.
Cohere Labs scraped 696K agent tools from 123K MCP servers and found only 2.6% can actually finish a real job task alone.
InclusionAI open-sources a 124B-parameter mixture-of-experts vision-language model with 5.5B active parameters, 256K context, and MIT license.
A new hybrid Structure-from-Motion framework from NAVER Labs unifies real-time SLAM and offline reconstruction, beating even calibrated systems while running uncalibrated.
A new CLI treats AGENTS.md as a set of weights, mining your agent transcripts to propose evidence-backed edits under a strict token budget.
TaichuAI released a 9B open-weight vision-language model that leads spatial reasoning and agent benchmarks against much larger open and closed rivals.

Meta Superintelligence Labs' first image model lands at #4 for editing and #5 for text-to-image, undercutting Google's Nano Banana 2 by more than 6x on price.

Anthropic's ant CLI now lets you declare agents, skills, environments, memory stores, and deployments as files and reconcile them like Terraform.

SpaceXAI opens its always-on agent platform to companies with access, network, and audit controls, plus two weeks of free usage for existing customers.

Prime Intellect rebuilt weight sync on RDMA and vLLM tracing, dropping GLM-5.2's 1.6TB policy transfer from 86 seconds to under 4.
Hermes Desktop now auto-detects your hardware, picks a fitting local model, downloads it, and configures the runtime without any manual setup.

OpenAI's new flagship saturates ARC-AGI-3 and FrontierMath Tier 4, drives a browser at 1.9x speed, and crosses the Critical cybersecurity threshold.
webAI's 3.66B Granite-based model specializes in formal logic, running locally in 2 GB while beating much larger reasoning models on logic benchmarks.

Anthropic is proposing TypeScript Function Hooks for Claude Code, letting plugins deeply customize behavior with Express-style middleware and admin-controlled side effects.
Tencent Hunyuan's environment evolution builds off-policy lineages of increasingly hard terminal tasks, boosting Qwen3.6 agents by up to 18 points on Terminal-Bench.

Warp launches Factory Benchmarks, letting teams replay real agent runs to pick the best model per task and cut cost-per-PR from $80 to $30.

Google Research and HHMI Janelia released a wiring diagram of 166,000 neurons and 125 million synapses spanning the male fruit fly brain and nerve cord.
A new attention kernel unlocks Blackwell's FP4 tensor cores for inference, hitting 2.13x BF16 forward throughput on GB200 while training stays partly in FP8.
Keep the project on disk, treat lunch as a bill, and score the first finished tasks rather than dollars per million tokens

Runway's new general world model streams continuous 720p video and 48kHz audio in real time, taking text prompts and camera input as you play.