AI
LongCat-DeepResearch Technical Report
Publication: Hugging Face Daily Papers
Author: Not available in collected metadata
Source published: 27/09/2026, 20:00:00 UTC
Briefing created: 30/09/2026, 04:46:55 UTC
Based on a feed excerpt; full article content has not been acquired. Source evidence retrieved 30/09/2026, 04:18:24 UTC.
Briefings
LongCat-DeepResearch is a system integrating an enhanced LongCat model with multi-agent workflows to generate evidence-grounded reports. It separates global planning from detailed investigation, coordinating section-level revisions. The system supports mid-training and post-training research task construction. Performance on DeepResearchBench is 55.25, on DeepResearchBench II is 51.35, and on ResearchRubrics is 79.83. On an in-house benchmark, it scored 76.04, ranking second among four compared systems. Development-set analyses show benefits from combining planning perspectives, while further planning refinement has mixed effects.
Key points
- LongCat-DeepResearch combines an enhanced LongCat model with a multi-agent workflow to produce evidence-grounded reports by separating global planning from detailed investigation.
- The system supports construction of research tasks and trajectories for the mid-training and post-training of LongCat's general-purpose models.
Context and limitations
Coverage is limited to the provided source evidence, which includes specific benchmark scores and performance metrics but lacks details on implementation costs, user adoption rates, or long-term stability.
Generated by Bittrees-hosted qwen35-2b sha256:aaf42c8b7c3c on 30/09/2026, 04:46:55 UTC. Based on publisher-provided feed evidence; may not cover the full original article. Check the source for details.
View archived briefing on IPFS
bafybeicygujg3yyejl645b3djnzjuwsksxpawheycwroo7qijnxd7jx67i