3 ms·
Wandr Benchmark: Evaluating Research Agents That Must Search Wide and Deep
- tagawa 3mo agoRepo with benchmark tasks, evaluation harness, tech report: https://github.com/perplexityai/wandr https://github.com/perplexityai/wandr
loading thread…
3 ms·