Claude Code and Codex Score Zero Running DeepSeek on kw-bench, OpenCode Gets 0.54
A new benchmark ran one model, DeepSeek v4.1 Flash, through four harnesses for two hours each. Two independent harnesses scored about half marks and the two lab harnesses scored zero. We go through that result and Arize's engineering behind agents, then Genie 4, the agent credential and exfiltration papers, two Anthropic stories on training data, and a few smaller items.
kw-bench: a VAT reconstruction across three generations of SAP
Arize on Code Mode
Arize on how Alex is built
Spine: a subtask-tree context runtime
AgentBaton: mid-task model handoff
DeepMind's Genie 4
Two Sigma on agent…
The skinny
The skinny isn't ready yet — notes appear once the transcript is processed.