AI brief
AWS has introduced Strands, an open source agent harness that it says uses fewer tokens than competing tools. Strands claims near-parity with Claude Code and Codex on benchmarks, though the benchmarks covered only coding agents and were run by AWS itself.
Why it matters: If the token savings hold up, an open source agent harness from AWS could offer a cheaper alternative to rival coding agents, but the vendor-run comparison limits how much the results can be trusted.
Written by AI from The Register AI + ML's published text.
Source
“Strands claims near-parity with Claude Code and Codex on benchmarks, though it only raced coding agents and marked its own…”
Reporting from a publisher outside the organisation being covered.
Read original story ↗