ProductDocsBlogPricingEnterprise
Talk to usGet started
← Back to all articles
Topic Tag Archive

#onnx

1 published article tagged with #onnx.

Architecture & Infrastructure•6 min read

Memory Without an LLM in the Loop

Eliminating external language models from the memory write path using local zero-shot ONNX models, FastEmbed, and deterministic civil date resolution to build sovereign, cost-predictable agent memory.

#onnx#local-extraction#gliner#fastembed#zero-llm
Oct 22, 2026Read article: Memory Without an LLM in the Loop →

All Tags

#agent-memory (6)#memory-fundamentals (3)#source-honesty (2)#spector (2)#arousal (1)#associative-memory (1)#byok (1)#cognitive-parameters (1)#consolidation (1)#cryptographic-shredding (1)#data-governance (1)#dream-pathway (1)#fastembed (1)#gdpr (1)#gliner (1)#hopfield-networks (1)#local-extraction (1)#mf-001 (1)#onnx (1)#pattern-completion (1)#roadmap (1)#valence (1)#verifiable-erasure (1)#zero-llm (1)

Memory that remembers what matters, and forgets the rest.

Product

  • How it works
  • Pricing
  • Enterprise
  • Benchmarks

Developers

  • Quickstart
  • SDKs
  • API reference
  • Redoc
  • Swagger UI

Company

  • Spectrayan
  • Blog
  • Contact
  • support@ebbmem.ai
© 2026 Spectrayan. ebbmem is a Spectrayan product.