Repository navigation
docs: host multi-agent orchestration benchmark chart #479
Description
Multi-Agent Orchestration Benchmark chart for the English and Chinese Raven README.
Hermes, Claude Code, and Raven are compared under Qwen3.8-27B and DS-V4-Flash-0731. The chart reports Node F1, Edge F1, Partial Order Accuracy, and Exact Match Rate on shared linear 0-1 scales. All 24 supplied values retain three decimal places. RavenX (Ours) is displayed as Raven (Ours).
Activity
docs: refresh bilingual readmes and agent benchmark figures (#480)
LivXue commentedon Sep 23, 2026
Recompressed copies of the README assets hosted in this thread, for README.md and README.zh-CN.md. Every still is now under 400 KB. The old comments stay as they are; each item names the asset it replaces and how it was re-encoded.
replaces
e333694a-0f4c-4f27-8bfe-8120ff5339a0: 426 KB -> 289 KB, lossless PNG recompression, identical pixelsdocs: recompress the README images over 400 KB and retitle the showca…
docs: refresh bilingual readmes and agent benchmark figures (#480)
docs: recompress the README images over 400 KB and retitle the showca…