
holaOS Ships Beta 0.1 Giving AI Agents Persistent Memory Across Workstreams
holaOS Beta 0.1 ships persistent multi-workspace agents with shared memory, sub-agents, and 100+ integrations — open-source and free to start
20 stories

holaOS Beta 0.1 ships persistent multi-workspace agents with shared memory, sub-agents, and 100+ integrations — open-source and free to start

Community quantizations by RealRebelAI shrink Alibaba's 30 GB character animation model into GGUF files that fit on 8-12 GB consumer cards.

Pythia bundles 48 keyless live feeds into one local JSON payload, giving agents planetary situational awareness plus multi-horizon forecasts through Ollama.

A new framework exposes a critical blind spot in reward-hacking monitors: the synthetic data used to train them doesn't reflect how models actually cheat during RL.

A developer squeezed a 28.9M parameter language model onto an $8 ESP32-S3 chip by moving most weights into flash memory.

Ant Group's Ling 3.0 Flash packs 124B parameters with only 5B active at inference, beating its own 1T flagship on most benchmarks at a fraction of the compute cost.

Anthropic ships four Managed Agents upgrades: spend budgets, inference geo-pinning, repo-loaded skills, and mid-session advisor models

Claude Code v2.1.224 lets independent sessions talk to each other directly, eliminating the need to manually relay context between terminals
Why agents fail even when the right rule stays in context

OpenAI's upcoming Astra model has crossed a new internal safety threshold for cybersecurity, triggering mandatory development controls and a new era for AI safety governance.

Anthropic's auto mode beats human review on every safety metric, and becomes the default in Claude Code on August 14 for Pro, Max, and Team plans

Prime Intellect's verifiers 0.3.0 and prime-rl 0.8.0 let you define, run, and train multi-agent RL environments with composable Agent and Env abstractions

Artificial Analysis rebuilds its image arena around 10 real-world use cases and 9 model capabilities, revealing where GPT Image 2's dominance ends and cheaper models take over
A 35B open model post-trained on self-generated adversarial tasks matches trillion-parameter systems on frontier science and coding benchmarks, now packaged as GGUF for local inference.
Suno's Voices feature, which lets you sing on AI-generated tracks using your own cloned voice, is now available natively on iOS and Android

Mem0 ships a rewritten memory algorithm with +27pt accuracy gains, single-pass extraction, entity linking, and temporal reasoning — at 3-4x lower token cost

MiniMax dropped an 11B open-weight music model that generates full five-minute songs with vocals from lyrics and a caption, in one shot.

MiniMax Code 2.0 launches with a rebuilt Pi Agent core, phone remote control, integrated browser, dual modes, and bring-your-own-key support

Krea adds ByteDance's Seedance 2.5 to its multi-model video platform, bringing 30-second single-take generation and 50 reference inputs to its creative workspace.

OpenAI's TypeScript Agents SDK gains Sandbox Agents in beta, giving JS/TS agents a real filesystem, shell access, and resumable workspace state