Figure Helix
Humanoid VLA model: a 7B VLM (System 2, 7–9 Hz) continuously updates a shared latent vector of behavioral intent that an 80M visuomotor policy (System 1, 200 Hz) consumes alongside the latest observation; runs onboard Figure robots.
Active · Verified 2026-10-22 · fresh as of 2026-10-22 · 31 cited facts, 2 with a third-party source · 1 recorded change · Suggest a correction
At a glance
- Type
- Temporal vision memory + tactile buffer vendor fix
- Tier
- T1
- Created
- 2022 (Figure AI founded May 2022 by Brett Adcock; Sunnyvale CA) third-party fix
- Latest release
- not applicable — not OSS fix
- License
- not applicable — not OSS fix
- GitHub
- not applicable — no GitHub repo fix
- Pricing
- searched not found fix
- Freshness
- 2026-10-22 fix
- Layer
- agent-infra repo fix
- Licensing
- commercial vendor fix
Memory 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
Alternatives to Figure Helix → live options ranked by shared design, licence, hosting and MCP support
Similar systems
Other vertical / domain-specific ai memory in the catalog, ranked by inbound references.
- Inworld AI T1
Formerly an NPC character engine with named Long-Term Memory; now sells realtime voice models (TTS-2, STT-1, speech-to-speech) and OpenAI-compatible open-model inference. Character/NPC memory is no longer the headline product.
- 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
Taken private: merged into Osaka Holdings (renamed BenevolentAI S.A.) and delisted from Euronext Amsterdam 2025-03-13. 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. As of Oct 2026 benevolent.com leads with Qor, powered by BenevolentAI, a life-science intelligence service for biopharma decision makers and consultancies. Qor (qor.one) is a continuously updated decision-intelligence service built on BenevolentAI's knowledge graph and reasoning methods: it tracks evidence on an asset or indication (competitive position, rival readouts, comparable deals), source-verifies each claim and flags when an earlier conclusion stops holding. It targets biotech teams and life-science consultancies.
- Causaly T1
Agentic AI platform for life sciences: orchestrated scientific agents over a curated biomedical knowledge graph (plus Bio Graph and Pipeline Graph), an evidence-retrieval system and an enterprise data fabric over literature, trials and patents.
Sources (30 cited facts, 2 third-party)
Each fact links to the page it came from; the badge says who published it (how to read sources).
- What it is Temporal vision memory + tactile buffer vendor
- What it does & what's distinct Humanoid VLA model: a 7B VLM (System 2, 7–9 Hz) continuously updates a shared latent vec… vendor
- Claims / benefits SOTA on humanoid logistics benchmarks (Scaling Helix paper). vendor
- Modalities vision + sensor + text vendor
- Created 2022 (Figure AI founded May 2022 by Brett Adcock; Sunnyvale CA) third-party
- Mindshare HN high: multiple top-100 posts vendor
- Funding Series C ›$1B at $39B post-money led by Parkway Venture Capital (2025-09-16) vendor
- Customers / scale 201-500 ppl BMW (11-month deployment - 30000 cars produced) vendor
- Data handling Onboard GPU inference; robot performance logs feed training data flywheel vendor
- Deployment Managed hardware deployment (robots at partner sites) vendor
- HQ US (Sunnyvale CA) third-party
- Founders / pedigree Figure AI; founder Brett Adcock (ex-Vettery CEO acquired by Adecco; ex-Archer Aviation f… vendor
- Memory volume / scale no scale claim (500 hours of human demos) vendor
- Performance Figure Helix: VLA model trained end-to-end on humanoid manipulation; vendor demos show p… vendor
- Versioning model-version semantics (policy checkpoints) vendor
- Namespace primitives robot-per-instance (per-robot memory) vendor
- Contradiction handling always overwrite (last-write-wins) vendor
- Forgetting policy episode-based memory; long-term memory module retains landmarks; episodic memory pruned … vendor
- Validated verticals robotics (humanoid, manipulation), autonomous-driving, embodied AI research vendor
- Time to running weeks-to-months (custom robot/vehicle integration) or research-only code release vendor
- Anti-fit not for non-embodied / non-physical use cases vendor
- Optimised for real-time perception + spatial reasoning + multi-modal sensor fusion vendor
- Adjacent infrastructure requires robot / vehicle hardware + sensor stack vendor
- Pros First major humanoid platform with explicit episodic + semantic memory layers in a publi… vendor
- Cons Robot-specific (Figure platform); production deployment is years away; published memory … vendor
- Project & sources figure.ai/helix vendor
- architecture-deployment-locus local vendor
- licensing-bucket commercial vendor
- positioning-platform-vs-product product vendor
- taxonomy-three-layer-rollup agent-infra repo
What changed
Changes to this record's lifecycle, licence, release, pricing and hosting facts, from the catalog's git history. Subscribe (Atom)