AI Agents & Infrastructure
warm-rust
Hybrid search, graph indexing, and vector-symbolic memory for AI agents
- 0community votes
- 26DR by Ahrefs no change since last check
- 2026founded
Ahrefs Domain Rating signal 13 Nov 2026, signal shifted 1 Nov 2026.
An embeddable memory engine for AI agents that runs locally on your machine with hybrid search, encryption at rest, and signed provenance.
What it is

WARM is a local-first memory and search engine for AI agents, built as a Rust library that runs inside your application without making network calls or sending telemetry. It combines hybrid search with graph-indexed vector retrieval and vector-symbolic fact memory, where bound facts are superposed into fixed-size vectors so every part carries information about the whole.
The engine provides AES-256-GCM encryption at rest, signed audit logs using Ed25519 with W3C DID identity, and a purge function that rewrites the journal and overwrites deleted data with zeros. It achieves 2.29-2.75x higher throughput than comparable open-source libraries on standard benchmarks while maintaining the same search quality.
WARM integrates with multiple languages and frameworks through native Rust bindings, Node.js npm modules, Python HTTP clients, and C ABI bindings for Go, Java and .NET. A local MCP server option enables integration with editors and AI agents, while WARM Cloud provides encrypted sync, backup and verification across devices without ever holding encryption keys or document contents.
Systems
- Hybrid search combining BM25 lexical and vector scoring
- Vertex graph vector index with 2.29-2.75x throughput vs hnswlib
- Vector-symbolic fact memory with holographic reduced representations
- AES-256-GCM encryption at rest with Argon2id key derivation
- Signed audit log with Ed25519 and W3C DID identity
- Secure purge that overwrites deleted data with zeros
- Crash-safe append-only log with checksumming
- Native Rust API with Python HTTP client and Node.js bindings
- MCP server integration for editors and AI agents
- WARM Cloud sync, backup and verification without server access to keys or documents
Pricing
Pricing model: Open source
Open source under Apache-2.0 license. WARM Cloud is in development with usage-based pricing for sync, backup and verification; details to be published at launch.
Want to see what warm-rust does next? Follow them for milestones, new products and stories by email.
warm-rust metrics
Signal readout for writerslogic.com: Ahrefs Domain Rating and Majestic link signals, arrows for the change since the last scan.
- 26 Domain Rating no change since last check
- 12 Trust Flow no change since last check
- 24 Citation Flow no change since last check
- 56 Referring domains no change since last check
- 358 External backlinks no change since last check
Ahrefs Domain Rating signal 13 Nov 2026, signal shifted 1 Nov 2026. Strongest signal topic: Computers/Software/Globalization (11). Majestic signal 1 Nov 2026, signal shifted 1 Nov 2026. Link signals from Majestic. Trust Flow, Citation Flow and Topical Trust Flow are trademarks of Majestic-12 Ltd.
warm-rust domain signal
Show the signal details
Public RDAP record, last read 12 Nov 2026, 09:39 UTC.
- Domain age 1 year
- Registered
- Expires
- Registrar NameCheap, Inc.
- DNSSEC Signed
warm-rust post-launch signal
The record for warm-rust after its launch on Nexora Launch (1 November 2026), from the readings we keep.
- 26 Domain Rating (Ahrefs) no change since launch
at launch: 26 (1 Nov 2026); latest: 13 Nov 2026 - 12 Trust Flow (Majestic)
first measured 1 Nov 2026 - 56 Referring domains (Majestic)
first measured 1 Nov 2026
Day 11 since launch. Next receipt: day 30, 1 Dec 2026. A change here is a reading, not a claim about its cause; lots of things move these figures. Anything we do not measure is not shown.
Got a launch coming up?
Plan your launch with our free playbook, from preparing your listing to choosing more launch directories.
Quick answers
- How is WARM different from Mem0, Zep, Chroma or Qdrant?
- WARM is a library you embed directly, not a hosted service you call. It includes encryption at rest, signed provenance, and text-erasing purge built into the open-source engine. Vector search runs on a vertex graph index that served 2.29x the queries per second of hnswlib at recall 0.95 on the same machine.
- Does it send my data anywhere?
- No. The engine makes no network calls and sends no telemetry. Embeddings are either supplied by you or computed by local models from weights you provide; it never downloads a model.
- How is this different from Mem0, Zep, Chroma or Qdrant?
- WARM is a library you embed, not a service you call, and it puts encryption at rest, signed provenance and a text-erasing purge in the open-source engine itself. Vector search runs on a vertex graph index; on SIFT-1M it served 2.29x the queries per second of hnswlib at recall 0.95 on the same machine (benchmarks section).
- What does WARM stand for?
- WARM stands for WritersLogic Adaptive Retention & Memory. The fact memory uses holographic reduced representations: many bound facts are superposed into one fixed-size vector, so every part carries a little of the whole and damage degrades recall gradually. Document search is conventional BM25 and cosine and does not claim that property.
- Is it really deleted when I delete it?
- Delete removes a record from every query at once; its journal frame stays until compaction or purge. Purge rewrites the journal and overwrites the old generation with zeros before it returns. Copies outside the store's files, such as backups, snapshots and SSD wear levelling, are out of any library's reach; encrypt the store when those matter. Signed deletion certificates are research, not yet in the shipped API.
warm-rust has a free listing. The owner can unlock the full profile with a paid launch for $5:
- Founder story (locked)
- Milestones (locked)
- Demo video (locked)
- Dofollow website link (locked)
Also on the frontier
-
Rank 1: ActionBox
Human approval infrastructure for AI agents and automation workflows
DR 5 no change since last check 0 votes. Log in to send ActionBox a vote -
Rank 2: Threadless AI
Hosted memory for AI agents that forgets like people do
DR 0 no change since last check 0 votes. Log in to send Threadless AI a vote -
DR 23 no change since last check 0 votes. Log in to send RELAI a vote -
DR 0 no change since last check 0 votes. Log in to send AgentHub a vote