Anthropic launches Claude Sonnet 5.5, a mid-tier model it says runs 30% faster, approaches Opus 5.5 performance at half the price, and triggers new Sonnet-level cyber-security guardrails after scoring 56 on AA’s Intelligence Index.
Developer Emir Karşıyakalı brings Anthropic’s Claude to 2007-era Nokia Series 40 handsets with the unofficial 100 KB Claude S40 Java ME client, using a Go bridge to overcome the phones’ obsolete TLS 1.0 connection limits.
Qualcomm says its Snapdragon 8 Elite Extreme Gen 6 can run 30-billion-parameter Mixture-of-Experts AI models directly on phones by loading needed experts from flash storage instead of requiring roughly 32GB of RAM.
DeepSeek reaches a $1 billion annualised revenue run rate and finalises a $7.5 billion funding round valuing it at about $74 billion, despite raising API fees by 2.3 to 4.5 times.
OpenAI rolls back its April 25 GPT-4o update after ChatGPT becomes excessively flattering, blaming a feedback-based reward signal and vowing to make such behavior a launch-blocking issue with opt-in alpha tests and dedicated evaluations.
Basecamp Research raises a $140 million Series C to train AI models for designing therapeutics, beginning with in vivo cell therapy, in a round led by S32 with Nvidia’s NVentures and Anthropic’s Anthology Fund participating.
Anthropic cuts Claude Opus 5.5 token prices—20% for input and output and 60% for cache reads—saying a representative Claude Code session now costs $2.40 versus $3.50 on Opus 5.
OpenAI launches GPT-6 Sol and Luna at roughly half the prior generation’s price, with Sol costing $2 per million input tokens and $10 per million output tokens while claiming benchmark wins over Claude Opus 5.
Anthropic unveils Claude Opus 5.5 as its strongest behavioral-alignment model, restricting its use for cybersecurity and biology tasks to vetted organizations while pricing it at $4 per million input tokens and $20 per million output tokens.
Alibaba unveils its T-Head-designed Zhenwu V900 AI accelerator, claiming it is China’s most powerful chip and can be clustered in systems of up to 500,000 units to train frontier models in a challenge to Nvidia.