Score8.6
Popularity763.0
Risklow
TierGold
Score breakdown
Usefulness9.0
Novelty10.0
Momentum10.0
Maturity9.2
Open-source/build7.4
Evidence7.2
Workflow potential9.7
Setup ease6.5
Popularity is tracked separately. Support, ads, sponsorships, and tips never affect these signals.
Why it matters
Useful for ML researchers studying long-horizon search agents, applied-AI teams training or fine-tuning stateful retrieval agents on SEC / BrowseComp+ style workloads, builders of RL-on-search pipelines who want a published 20B checkpoint as a comparison point, Tinker and vLLM users who want a GPT-OSS-compatible 20B search-agent to evaluate locally, and any operator who needs a stateful retrieval
Who should use it
Who should skip it
Skip if the source link, docs, or setup requirements do not match your workflow.
Risk explanation
Risk label needs manual review.