view article Article Feather DB on LongMemEval: embedded retrieval beats full-context GPT-4o for $2.40 8 days ago
view post Post 105 We ran Feather DB v0.8.0 on LongMemEval (ICLR 2025) — 500 questions across real multi-session conversations, up to 115K tokens each.**Score: 0.693** · GPT-4o full-context baseline: 0.640Full 500-question run with Gemini-Flash: **$2.40**Per-axis breakdown:→ Info-extraction: **0.942**→ Knowledge-update: **0.714**→ Multi-session: **0.606**→ Temporal: **0.477** ← the hard one, Phase 9 addresses thisArchitecture: Hybrid BM25+dense · adaptive temporal decay · embedded (no server) · p50 = 0.19ms · MITpip install feather-dbRaw results + audit JSONs: Hawky-ai/longmemeval-results See translation Reply
Sleeping Agents Feather DB — Living Context Engine 🪶 Search and explore product intel using semantic graph queries
Sleeping Agents clawID — AI Asset Watermarking 🐾 Invisible & visible watermarking for AI-generated images
Sleeping Agents clawID — AI Asset Watermarking 🐾 Invisible & visible watermarking for AI-generated images
HauhauCS/Qwen3.5-35B-A3B-Uncensored-HauhauCS-Aggressive Image-Text-to-Text • 35B • Updated Apr 5 • 1.11M • 1.38k
pm-AGI Collection The first open-source LLM benchmark for Performance Marketing — evaluating Meta Ads, Google Ads, critical thinking, and real-world action-based scena • 2 items • Updated Mar 13