Fresh stories
Report: Claude Mythos reportedly solves Erdős problem #90 in air-gapped test
Anthropic staff and outside observers said a Mythos-powered Claude Code setup solved Erdős problem #90 in an internet-blocked test. The result is still based on harnessed runs and social-thread disclosures, so watch for fuller verification before treating it as settled.
OpenRouter raises $113M Series B as weekly volume hits 25T tokens
OpenRouter announced a $113M Series B led by CapitalG and said weekly routed volume grew from 5T to 25T tokens in six months. The funding matters because the company is pitching itself as production infrastructure for multi-model deployments, not just an API convenience layer.
Weights & Biases launches MCP server with 20 tools for schema-first queries
Weights & Biases released an MCP server that exposes experiment data to Claude Code, Cursor, Codex, Gemini CLI, and Le Chat. The schema-first design helps agents inspect available metrics before pulling rows, which can prevent preview runs from overflowing context windows.
Report: Claude Mythos reportedly solves Erdős problem #90 in air-gapped test
Anthropic staff and outside observers said a Mythos-powered Claude Code setup solved Erdős problem #90 in an internet-blocked test. The result is still based on harnessed runs and social-thread disclosures, so watch for fuller verification before treating it as settled.
Qwen3.7 Max ships implicit caching for no-setup context reuse
Alibaba rolled out implicit caching for Qwen3.7 Max, automatically reusing repeated context without user setup. The update also lands with fresh benchmark results and broader coding-agent support across OpenCode and Hermes Agent.
Claude Code ships security-guidance plugin with repo-level claude-security-guidance.md rules
Anthropic added a security plugin to the Claude Code marketplace and said internal use cut security-related PR comments by 30-40%. Teams can use it to enforce repo or MDM-distributed policies before human review.
OpenRouter raises $113M Series B as weekly volume hits 25T tokens
OpenRouter announced a $113M Series B led by CapitalG and said weekly routed volume grew from 5T to 25T tokens in six months. The funding matters because the company is pitching itself as production infrastructure for multi-model deployments, not just an API convenience layer.
Grok Build Beta adds Toad and Kilo Code integrations plus a web Build tab
SynthID adds OpenAI, ElevenLabs, and Kakao partners as Search and Chrome gain verification
Warp Agent adds OpenRouter URLs and /model aliases for custom endpoints
Weights & Biases launches MCP server with 20 tools for schema-first queries
Top storiesthis week
Huawei pitches τ scaling with LogicFolding and a 1.4nm-equivalent 2031 target
Huawei outlined a τ scaling framework and LogicFolding design that shifts chip progress from node shrinkage toward shorter signal delay. The proposal matters because it targets performance, density, and yield gains without relying only on EUV-era process shrinks.
Microsoft benchmarks SkillOpt at +24.8 Codex points by editing skills, not weights
Microsoft Research released SkillOpt, which optimizes external skill files instead of fine-tuning model weights and reports best-or-tied results across 52 evaluation cells. The method matters because it improved Codex and Claude Code accuracy without extra inference-time calls.
Developers ship Chrome MCP, repo-graph search, and token compression for Claude Code and Codex
Independent developers released browser-control MCP tooling, repo-context graphing and packaging utilities, and token-compression helpers for coding agents. The cluster matters because agent workflows are now adding browser control, context packing, and cost controls as external infrastructure instead of waiting on raw model upgrades alone.
Developers compare 128GB workstations, M5 Max laptops, and 20/80 local-cloud agent splits
Developers published new local-first agent setups spanning 128GB workstations, M5 Max laptops, local-model checkers, and 20/80 local-cloud splits. The pattern matters because teams are moving extraction, coordination, and offline tasks off frontier APIs while keeping harder reasoning in the cloud.
Researchers and builders ship external memory layers with recipe stores and 33% cheaper updates
A new MeMo paper and several community memory systems converged on keeping knowledge outside the base model through recipe files, semantic and autobiographical stores, and background reconsolidation. The pattern matters because engineers are treating context loss as a systems problem instead of only asking for larger context windows.