Figure Helix

https://www.figure.ai/helix

System 2 (7B VLM) feeds tokens to System 1 (80M visuomotor policy at 200 Hz). Temporal memory module maintains visual-state history; force-feedback buffered as tactile memory proxy. 500 hours of human demos. Deployed in BMW logistics.

At a glance

Type
Temporal vision memory + tactile buffer
Tier
T1
Created
2022 (Figure AI founded May 2022 by Brett Adcock; Sunnyvale CA)
Latest release
not applicable — not OSS
License
not applicable — not OSS
GitHub
not applicable — no GitHub repo
Pricing
searched not found
Funding
$675M total $2.6B val Series B · 2024-02

Taxonomy

storage
vector
retrieval
attention
persistence
session
update
overwrite
unit
episode
governance
opaque
conflict
overwrite

When to use

Optimised for: real-time perception + spatial reasoning + multi-modal sensor fusion

Anti-fit: not for non-embodied / non-physical use cases

Pros & cons

Pros

First major humanoid platform with explicit episodic + semantic memory layers in a published architecture; backed by deep funding.

Cons

Robot-specific (Figure platform); production deployment is years away; published memory details are still high-level.

Claims & capabilities

SOTA on humanoid logistics benchmarks (Scaling Helix paper).

Technical surface

API surface
searched not found
Backend storage
searched not found
Deployment
Managed hardware deployment (robots at partner sites)
Embedding model
searched not found
Multi-tenancy
searched not found
MCP
no MCP support advertised — vertical product, no MCP server / client integration documented
A2A
no A2A protocol support advertised — vertical product, no A2A integration documented
OpenTelemetry
no OpenTelemetry integration advertised — vendor logs/observability not publicly documented

Similar systems

Other vertical / domain-specific ai memory in the catalog, ranked by inbound references.

  • NVIDIA ReMEmbR T3

    Builds long-horizon memory by captioning video segments with VILA, storing captions with timestamps + 3D position coordinates in MilvusDB. At query time, LLM iterates retrieval across text, time, and position modalities. Deployed on Nova Carter robot (Jetson Orin).

  • Abridge T1

    Clinician-assist ambient documentation. Source mapping: every AI-generated summary element traced back to the source utterance. Audit-and-trust layer over episodic memory. Built on proprietary 1.5M+ medical-encounter dataset.

  • ASAPP GenerativeAgent T1

    Treats memory as first-class architecture. Captures the digital footprint of every interaction; retrieves preference and history at engagement time. Public example: airline knowing a frequent flyer wants aisle seats with her son — preference-aware, not just history-lookup.

  • BenevolentAI T1

    Target identification / drug repurposing / mechanism tracing. 85+ data sources, petabyte-scale, rebuilt every few weeks. Wet-lab results re-enter the graph and shift downstream predictions — institutional experimental memory closing a feedback loop.

  • Causaly T1

    Drug discovery / target identification / causal mechanism tracing. The graph is the memory: 7 years of curated biomedical cause-effect relationships compounding with each new ingestion. Scientific RAG retrieves from a versioned causal substrate.

  • Character.ai T1

    Chat Memories (user-defined facts), auto-memories for c.ai+ subscribers, pinned memories, in-context retention. PipSqueak 2 model (April 2026) reduces in-conversation drift. Memory Visualization meter forthcoming.

Row last verified 2026-05-14. Catalog data is CC-BY-4.0 — see how to read this.