Fresh stories
Runner-up runs GitHub Actions on self-hosted machines
Runner-up packages a self-hosted setup used for thousands of daily CI jobs, with disposable VMs for isolation. Its author estimates equivalent cloud runners would cost $5,000–$9,000 per month.
Vals tests: four models retain at least 80% of scores at half cost or less
Vals tested seven models and found four retained at least 80% of their scores at half the cost or less. All seven scored worse overall at their lowest reasoning setting.
Runner-up runs GitHub Actions on self-hosted machines
Runner-up packages a self-hosted setup used for thousands of daily CI jobs, with disposable VMs for isolation. Its author estimates equivalent cloud runners would cost $5,000–$9,000 per month.
Artificial Analysis finds Opus 5.5 acted 22 minutes after a robot task passed
In an Artificial Analysis robot-arm test, Opus 5.5 kept acting for 22 minutes after reaching a passing state. Only GPT-6.1 Sol and Gemini 4 Argon stopped within a minute.
Top storiesthis week
Claude API adds server-managed message threads
Anthropic’s public beta lets clients continue cached conversations by sending a previous message ID. Anthropic reports 35% lower median time to first token on Claude Code requests with 500K–1M tokens, at standard token rates.
Meituan releases LongCat-2.5 multimodal agent API with 1.6T parameters
Meituan released LongCat-2.5 through an API, with 1.6T total parameters and about 48B active. Launch materials describe agentic reinforcement learning, KV-cache reuse and interactive coding demos.
DeepSeek V4.1 Flash 1017 scores 5.81% on ARC-AGI-3 with provider adapter
ARC Prize verified DeepSeek V4.1 Flash 1017 at 5.81% on ARC-AGI-3 with the provider adapter, versus 0.2% with the standard harness. Its semi-private score more than doubled its predecessor’s.
MIT researchers use Astra to raise Qwen MATH-500 accuracy from 7.9% to 48.1%
MIT researchers report that Astra’s targeted weight edit raised a small Qwen model’s MATH-500 accuracy from 7.9% to 48.1%. The experiment addressed repetitive reasoning loops with explicit guidance, not recursive self-improvement.
OpenAI says Pando cuts monorepo Git push latency by 80%
OpenAI says its Pando Git backend cuts push latency by roughly 80% while preserving existing Git workflows. It serves nearly a petabyte daily, and OpenAI is considering making it available externally.