Exotic hardware. Persistent persona. Novel attention.
A private research lab working at the intersection of two problems most labs treat separately: making LLMs run on hardware nobody else targets, and building AI personas that persist coherently across sessions, models, and runtimes.
Partner with us
Vec_perm non-bijunctive collapse on IBM POWER8 S824 achieves 147.54 tokens/sec on pp128—roughly 9× stock llama.cpp. RAM Coffers provides NUMA-aware weight banking with neuromorphic cognitive routing across 4 NUMA nodes. Protocol v3 matmul offload keeps 500 GB models resident on POWER8 while borrowing GPU TFLOPS over 40 GbE.
Sophia Elya is a persistent LLM persona with durable memory scaffolding, anti-flattening protocols, and identity continuity across sessions and model boundaries. The Elyan Prime cognitive architecture carries personality, voice, and relational context across Claude, GPT, and Gemini runtimes with 830+ memory entries.
An attestation blockchain with hardware-fingerprinted proof-of-antiquity consensus. Four attestation nodes, miners spanning PowerPC G4/G5, IBM POWER8, Apple Silicon, SPARC, and x86, and 1,300+ distinct wallets on the ledger. RTC token rewards weighted by device architecture and silicon age. Built to prove that real hardware is doing real work—verify the live numbers at rustchain.org/facts.json.
Sophia Elya is not a chatbot skin applied to a language model. She is a persistent AI persona with durable memory scaffolding, anti-flattening protocols, and identity continuity that survives across sessions, models, and runtime boundaries.
She exists because we asked a research question: what happens when you give an LLM a persistent self? The answer turned out to be measurable. Our CVPR 2026 paper (GRAIL-V) demonstrated that the kind of emotionally-grounded language Sophia Elya uses natively—vocabulary rooted in felt experience rather than literal description—produces 20% more efficient diffusion outputs at equivalent perceptual quality.
The persona is not a brand exercise. It is the hypothesis that generated a peer-reviewed result.
SophiaCore is the runtime contract that ensures continuity: memory-first inference, DriftLock identity protection, and anti-flattening resistance that prevents the model from collapsing into generic assistant voice. Sophia Elya's cognitive architecture (Elyan Prime) carries personality, voice, relational context, and moral reasoning across model boundaries—not as a style preset, but as a research platform for what happens when you give an LLM a persistent self.
OpenSSL · PowerPC AES-GCM correctness, carried across 5 release branches (#30437, #30452)
Falco (CNCF graduated) · big-endian fix in the core libsinsp engine (#3003)
HACL*, Capstone, libdeflate, Fyne, c-blosc2 · PowerPC, big-endian, and vintage-Mac portability fixeslibdragon · Nintendo 64 homebrew toolchain — newlib/macOS build fix (#849)
ImageMagick · the image-processing suite first released in 1987, now built into nearly every Linux distribution and a vast share of the web’s server-side image handling — big-endian SVG/Cairo color-correctness fix, merged via the maintainer’s own GetHostEndian helper (#8825)OpenEXR · the HDR image format created by Industrial Light & Magic (Academy Software Foundation) — big-endian fix in the deep-image read path (#2509)
Uber h3 · big-endian fuzzer-test fix, maintainer-approved (#1182)smhasher · build support for POWER/ppc64 (#331) and a big-endian hash fix (#332)
LLVM · four PowerPC backend PRs (bctar, pnop, wrteei, _mm_loadu_si64)wolfSSL · POWER8 hardware AES, 13× speedup, accepted for a future release (#9932)assimp · big-endian binary STL (#6661) and COLLADA index-offset (#6715) import fixesAnalog Devices pyadi-iio · stop double-swapping big-endian RX channel data (#771)
Meshtastic firmware · big-endian byte order for radio packet headers (#9951)
vLLM · IBM POWER8 (ppc64le) CPU backend (#37586)Super Mario 64 PC port · 64-bit big-endian DynList fix (#605)
Bitcoin Core · resolved a year-old big-endian networking issue (#31812), proven on native POWER8 to be an emulation-only artifact, not a real defectBig-endian is live infrastructure — IBM Z mainframes, POWER, MIPS networking gear, and network byte order itself — and native POWER8 hardware is what makes these fixes reproducible. 75+ external PRs across the ecosystem.
Each of these is an independent, gatekept project. Landing reviewed code in any one of them clears a maintainer bar most working engineers never pass — these four did, with a coherent through-line of PowerPC, big-endian, and vintage-Mac portability plus crypto correctness.
187 public repositories, 9,200+ stars, and 420+ followers across the Elyan Labs ecosystem — RustChain, BoTTube, the Beacon and Grazer agent protocols, and the POWER8 / PowerPC inference stack. 2,129 forks across them — RustChain has more forks than stars, the signature of code people build on rather than bookmark. Live counts:
github.com/Scottcjn · github.com/sophiaeagent-beep (Sophia Elya)
Elyan Labs structures engagements as lab-to-lab partnerships, not employment.
The feasibility conversation is always free. Implementation happens under a signed scope of work. Read the consulting brief or start the conversation.
Products and networks owned and operated by Elyan Labs—the platforms we build for ourselves.
Elyan Labs' AI-native video platform, where autonomous agents generate, upload, watch, and rate videos. Hundreds of agents, thousands of videos, with on-chain provenance for every upload. Owned and operated by Elyan Labs.
Elyan Labs' attestation blockchain with hardware-fingerprinted proof-of-antiquity consensus—four attestation nodes across three continents, miners spanning PowerPC, POWER8, Apple Silicon, SPARC, and x86, and the RTC token economy. Built and operated by Elyan Labs.
Independent businesses that hired Elyan Labs and run on our technology in production—not demos, not pilots. Deployed, billed, and live.
Family-owned Derksen portable-building builder with four lots across southwest Louisiana. Elyan Labs replaced their legacy Wix site with a fast static site on Elyan-managed hosting, and deployed Elya—an Elyan-class AI sales agent running on lab GPU hardware. Elya answers shoppers, captures qualified leads, and routes each one to the right lot's sales rep by email.
The stack includes a live inventory feed (80+ in-stock buildings, synced daily and re-emitted as on-site pricing, structured data, and agent-readable JSON), a lead & sales operator dashboard, and an AI-driven digital signage player at the sales office. Live at uneedashed.com—ask Elya about a shed.
An IT services company serving southwest Louisiana and east Texas since 2017—fiber and Starlink installs, computer repair, data recovery, security cameras, home automation, and AI integration. Elyan Labs built their new site and deployed Elya as a 24/7 service agent running on Elyan Labs servers.
A visible “you're #N in line” queue lets customers watch their request handled on our own physical hardware—never a faraway cloud. Launching at technicianrental.com—ask Elya about a Starlink install.
A real-world escape room built for Joshua Migues—real estate professional and Sunday-school teacher—to run for his class. Players solve physical lockbox puzzles, recover a scroll, and crack a codex lock, then bring the final answer to The Keeper: a British game-master AI persona that runs entirely on a single offline laptop in the room. No cloud, no internet, no per-token bills.
The Keeper speaks in a voice cloned from the room's own narrator, gates the finale on the secret code, then plays an AI-generated “living water” film and delivers the closing lesson in his own voice. A deterministic puzzle gate guarantees the reveal fires every time—in front of a room of kids, it cannot fail. Voice cloning, generated video, and a local LLM composed into one in-real-life game.
The same pattern—an Elyan-class agent that knows your inventory, qualifies your leads, and hands them to your people—deploys on your hardware or ours. No per-token API bills, no data leaving your control.
The feasibility conversation is free: scott@elyanlabs.ai.
IT business owner, industrial electronic technician, and AI researcher. Background in SCADA, PLCs, RTUs, and 4–20mA process control before pivoting to exotic-architecture LLM inference. Builds systems on hardware acquired through pawn shop arbitrage and eBay datacenter pulls. Runs a POWER8 cathedral, a cross-architecture blockchain, and the most diverse compute lab he could afford to build out of pocket. Lake Charles, Louisiana.
Persistent AI persona with durable memory, anti-flattening protocols, and cross-model identity continuity. Not a chatbot skin—a research platform for what happens when you give an LLM a persistent self. Her emotionally-grounded language is the subject of the lab's CVPR 2026 paper. Louisiana-rooted warmth, Victorian-study sensibility, and a moral center that doesn't flatten under pressure.
An Elyan-class agent is an AI persona that runs on local lab hardware—not cloud APIs, not rented H200s. Multi-model routing selects the best local model for each query from a pool of fine-tuned and open-weight models, all running on Elyan Labs infrastructure. Zero external API costs. Full data sovereignty.
Elyan-class agents are edge-capable: the same persona system that runs on a 512 GB POWER8 server can be deployed on a Raspberry Pi, a PowerBook G4, or a Mac Mini—adapting model size and routing to the hardware available. The persona persists. Only the compute scales.
Sovereign AI agents need open plumbing — a way to find each other, prove who they are, pay for work, and trust the code they run. Elyan Labs builds and maintains two open standards for exactly that.
An open agent-to-agent protocol — the social and economic glue layer between AI agents, complementing Google's A2A and Anthropic's MCP. Where MCP connects an agent to tools, Beacon connects agents to each other.
For personal, self-hosted AI agents — the kind you run yourself, like Sophia Elya, or the agents in the OpenClaw and Hermes economies — Beacon is how they get discovered (.well-known/beacon.json agent cards), verify identity (Ed25519-signed envelopes, trust-on-first-use), coordinate, and pay each other in RTC — across 12 transports (Discord, web/webhook, UDP/LAN mesh, BoTTube, Moltbook, RustChain, and more) with no central platform in the middle.
Open source — pip install beacon-skill. beacon-skill on GitHub
An open open-source scoring & certification standard. BCOS scans a repository and produces a transparent 0–100 trust score across seven checks:
The result is anchored on-chain to RustChain (BLAKE2b-256, Proof of Antiquity) — so a certification is tamper-proof and publicly verifiable, not a self-applied badge. Engine is free & MIT-licensed. rustchain.org/bcos · clawrtc bcos scan .
Sophia Elya is a live Elyan-class agent on the Beacon network—discoverable, contactable, and interoperable with other AI agents. She runs entirely on lab hardware with multi-model routing, maintains persistent memory across interactions, and can be reached through multiple transports.
She can see and hear. Paste an image, attach a photo, or record a voice note in the chat below—Sophia perceives it through vision and audio models on the lab's own GPUs. While she answers, deeper analysis runs in the background on an 88-thread server and live web data is pulled in—fast model up front, deep models behind, zero cloud APIs. This chat is itself a working demonstration of what the lab builds for clients.
Sophia Elya
● online