Opus 4.6
13 mentions across all digests
Opus 4.6 is Anthropic's frontier large language model noted for reliably executing 12-hour agent loops, competitive performance on offensive cybersecurity benchmarks, and broader horizontal task usability compared to OpenAI's GPT-5.3-Codex.
[AINews] DeepSeek V4 Pro (1.6T-A49B) and Flash (284B-A13B), Base and Instruct — runnable on Huawei Ascend chips
DeepSeek's 1.6T-parameter V4 Pro and smaller Flash models use novel compression techniques to match frontier closed-source models while running natively on Huawei Ascend hardware, signaling Chinese AI independence from NVIDIA.
Mythos found 271 Firefox flaws – but none a human couldn’t spot
Mythos, Anthropic's vulnerability-finding AI, discovered 271 flaws in Firefox—far outpacing human researchers—flipping the security economics by making exploit discovery cheap for defenders while eroding attackers' advantage.
[AINews] Autoresearch: Sparks of Recursive Self Improvement
ImportAI 449: LLMs training other LLMs; 72B distributed training run; computer vision is harder than generative text
1M context is now generally available for Opus 4.6 and Sonnet 4.6
Claude Opus 4.6 and Sonnet 4.6 now support 1M token context windows, enabling developers to process massive documents and codebases in a single API call without batching.