Honcho
Open-source memory infrastructure, powered by custom models
Details
- External ID
- 46781717
- Source
- HN
- Company
- —
- Product
- Honcho
- Website domain
- github.com
- Launched
- Jan. 27, 2026
- Cohort
- —
- Upvotes
- 8
- Upvotes percentile
- 0.41699604743083
- Tags
- —
- Fetched at
- Sept. 7, 2026, 9:25 p.m.
- Updated at
- Sept. 7, 2026, 9:25 p.m.
Description
Hey HN,It’s Vineeth from Plastic Labs. We've been building Honcho, an open-source memory library for stateful AI agents.Most memory systems are just vector search—store facts, retrieve facts, stuff into context. We took a different approach: memory as reasoning. (We talk about this a lot on our blog)We built Neuromancer, a model trained specifically for AI-native memory. Instead of naive fact extraction, Neuromancer does formal logical reasoning over conversations to build representations that evolve over time. Its both cheap ( $2/M tokens ingestion, unlimited retrieval), token efficient and SOTA: LongMem (90.4%), LoCoMo (89.9%), and BEAM. On BEAM 10M—which exceeds every model's context window—we hit 0.409 vs prior SOTA of 0.266, using 0.5% of context per query.Github: https://github.com/plastic-labs/honchoEvals: https://evals.honcho.devNeuromancer Model Card: https://plasticlabs.ai/neuromancer)Memory as Reasoning Approach: https://blog.plasticlabs.ai/blog/Memory-as-ReasoningRead more about our recent updates: https://blog.plasticlabs.ai/blog/Honcho-3Happy to answer questions about the architecture, benchmarks, or agent memory patterns in general
Enrichment
- Theme
- modular ai agent skills and toolkits
- Vertical
- Horizontal
- Function
- Model & infra
- Audience
- Developer
- AI stance
- AI-native
- Project type
- Commercial product
- Normalized one-liner
- memory infrastructure for ai applications
- Manually corrected
- False
Could you build this?
No Honcho relies on custom fine-tuned reasoning models and specialized stateful memory infrastructure rather than basic vector retrieval.
What it would actually take: Building Honcho requires fine-tuning specialized language models for dialectic reasoning, theory-of-mind tracking, and contextual extraction. The backend architecture involves high-throughput streaming state engines and relational/graph storage to maintain persistent multi-agent user representations. Doing this requires deep NLP research expertise, model distillation pipelines, and high-performance agent architecture design.
Discussion
No comments on this launch.
Competitors
Other products that read as similar to this one — 404 launches clear the similarity bar, closest 8 shown.
Attention rank: #228 of 405 (itself plus its competitors, highest first — normalized so YC and Product Hunt are compared fairly).
Launched 89 days after the earliest competitor.
- A file-based agent memory framework that works like skill · hn · 2026-01-06 · 11 upvotes · similarity 0.53
- ClawMem · hn · 2026-03-22 · 5 upvotes · similarity 0.49
- Memori · ph · 2026-05-28 · 168 upvotes · similarity 0.48
- retain-ai · github · 2026-09-11 · 8 upvotes · similarity 0.48
- MemoryOps · hn · 2026-06-22 · 5 upvotes · similarity 0.48
- Valori · ph · 2026-09-22 · 64 upvotes · similarity 0.47
- AI memory with biological decay (52% recall) · hn · 2026-04-26 · 98 upvotes · similarity 0.47
- Unified multimodal memory framework, without embeddings · hn · 2026-01-07 · 7 upvotes · similarity 0.47
Other launches for this product
- No other launches for this product.
Same idea, different domain
Nobody's really built a model & infra tool for Fintech yet.