Anthropic has launched Claude Haiku 5.5, a cost-efficient model featuring a 1M-token context window that ranks 27th in the Agent Arena leaderboard.
OpenAI has launched GPT-6 Sol and Luna with enhanced prompt caching that cuts input token costs by up to 90% and reduces latency for developers.
DeepSeek is hiring approximately 150 engineers to scale its backend infrastructure and has lowered off-peak API prices for its Flash series.
DeepSeek has started internal testing for the V4.1 Flash model, featuring a new architecture and high output speeds reaching 507 tokens/s.
