Hardware FixRecommendedDevice not working? Your driver may be the problemCheck updates for common hardware issues.Fix DriversHalloween DealsAmazon USHalloween week deals: useful picks before month-endAmazon US: current deals, useful picks and tech finds.Check DealsClean PCRecommendedOne scan can reveal what keeps slowing WindowsLook for cleanup and repair opportunities.Run Scan×
Skip to content
World desk4 min

I Built a 34-Agent AI Swarm on My Phone—Here’s How It Worked

A first-person account of a 34-role AI swarm on Android, its message-bus architecture, concurrency limits, and the caveats behind its performance figures.
Special offer. See more information about Outbyte and uninstall instructions. Please review EULA and Privacy policy.

Samuel James Hiotis says he built a 34-agent AI system on an Android phone by coordinating lightweight agent wrappers through Redis Streams, limiting how many ran at once, and sending their results to a synthesizer. The key distinction is that 34 was the number of agents he defined—not the number active simultaneously. His September 30, 2025 account describes four to six active agents at a time, with some work handled by local models and final synthesis eligible for hosted inference. The figures are his report, not an independently reproduced test.

What “34 agents” meant in this build

In Hiotis’s design, an agent was a lightweight wrapper around an inference call, not a separate always-running AI model. An orchestrator broke an incoming task into subtasks, dispatched them to agents, and collected their outputs for a synthesizer. Agents communicated through a shared message bus.

As an Amazon Associate I earn from qualifying purchases.

That arrangement let the author define many specialist roles without running every role at the same time. He reports keeping four to six agents active concurrently. The article’s central engineering idea is therefore task coordination under resource limits, rather than 34 models operating in parallel.

Special offer. See more information about Outbyte and uninstall instructions. Please review EULA and Privacy policy.

The reported phone and software stack

Hiotis says he built the system over three weeks using Android and Termux. His account names Python 3.11, Ollama, Redis Streams, Uvicorn, and a simple Flask interface. Small quantized models ran locally, while requests could be escalated to hosted models for final synthesis.

#1 Best Overall
Sale
Motorola Moto g - 2026 | Unlocked | Made for US 4/128GB | 50MP Camera | Pantone Slipstream, Cellular_Phone
  • Universal unlocked. Compatible with all major U.S. carriers, including Verizon, AT&T, T-Mobile and other prepaid carriers.
  • Super-bright, super-smooth 6.7" display. See your screen clearly even outdoors in sunlight, and enjoy seamless views with a fast-refreshing 120Hz display.*
  • AI-powered camera system. Take stunning photos in any light with the 50MP camera**, look your best with a 32MP selfie cam*****, and capture extreme close-ups.
  • Superfast 5G performance. Unleash your entertainment at 5G speed*** with the MediaTek Dimensity 6300 chipset and up to 12GB of RAM with RAM Boost****.
  • Long-lasting battery + TurboPower charging. Power through day after day with a 5200mAh battery, then get hours of power in just minutes.****
Part Role in the reported setup
Android with Termux Phone environment for running the build
Python 3.11 Implementation language for the agent wrappers and orchestration
Redis Streams Shared message bus for agent communication
Ollama and small quantized models Local inference, as described by the author
Hosted models Optional escalation path for final synthesis
Uvicorn and Flask Server and simple interface named in the account

The source does not identify the phone model or Android build, provide a repository, or give installation instructions detailed enough to reproduce the setup exactly. Treat this as an account of an implementation, not a tested recipe for a specific handset.

How a task moved through the swarm

  1. Decompose: The orchestrator turns the request into smaller tasks that can be assigned separately.
  2. Dispatch with a limit: It sends subtasks to available agents while a semaphore caps simultaneous work. Hiotis emphasizes that this limit matters for avoiding memory exhaustion on a phone.
  3. Collect and synthesize: The system gathers agent outputs and passes them to a synthesizer, which produces a combined result.
  4. Review and format: In the author’s RISC-V research-brief demonstration, three researchers handled subtasks, followed by a synthesizer, reviewer, and formatter.

Hiotis reports that the RISC-V brief took about 90 seconds. That is a self-reported demonstration time, not an independent benchmark. The presence of a reviewer in the workflow also does not establish that the resulting brief was factually accurate.

Rank #2
Sale
Samsung Galaxy A17 5G Smart Phone 128GB US 1 Yr Manufacturer Warranty Black | Large AMOLED, High-Res Camera, Durable Design, Super Fast Charging, Expandable Storage, Circle to Search, 2025
  • YOUR CONTENT, SUPER SMOOTH: The ultra-clear 6.7" FHD+ Super AMOLED display of Galaxy A17 5G helps bring your content to life, whether you're scrolling through recipes or video chatting with loved ones.¹
  • LIVE FAST. CHARGE FASTER: Focus more on the moment and less on your battery percentage with Galaxy A17 5G. Super Fast Charging powers up your battery so you can get back to life sooner.²
  • MEMORIES MADE PICTURE PERFECT: Capture every angle in stunning clarity, from wide family photos to close-ups of friends, with the triple-lens camera on Galaxy A17 5G.
  • NEED MORE STORAGE? WE HAVE YOU COVERED: With an improved 2TB of expandable storage, Galaxy A17 5G makes it easy to keep cherished photos, videos and important files readily accessible whenever you need them.³
  • BUILT TO LAST: With an improved IP54 rating, Galaxy A17 5G is even more durable than before.⁴ It’s built to resist splashes and dust and comes with a stronger yet slimmer Gorilla Glass Victus front and Glass Fiber Reinforced Polymer back.

Why concurrency and memory were constraints

Hiotis estimates about 200 MB of context per agent and multiplies that by 34 to arrive at 6.8 GB. He contrasts that estimate with a 6 GB phone, although he does not identify the device. Those are the author’s estimates, not measured memory figures; they help explain why the design activated only a subset of agents at once rather than proving a particular phone’s memory requirements.

Special offer. See more information about Outbyte and uninstall instructions. Please review EULA and Privacy policy.

He also reports an approximately 1.2 GB APK-less footprint, attributing most of it to model weights. The article does not specify the model configuration or measurement method, so this should not be treated as a generally expected install size.

Rank #3
Sale
Samsung Galaxy A16 5G 128GB Cell Phone, Unlocked Android Smartphone, Large AMOLED Display, Durable Design, Super Fast Charging, Expandable Storage, US Version, 2025, Blue Black (Renewed)
  • Charger NOT Included, 6.7" Super AMOLED FHD+, 90Hz Refresh Rate, 385 ppi, 800 nits (HBM), 1080x2340px, 5000mAh Battery
  • 128GB, 4GB RAM, microSDXC, Exynos 1330 (5nm), Octa-Core, Mali-G68 MP2 or Mali-G57 MC2 GPU
  • Rear Camera: 50MP, f/1.8 (wide) + 5MP, f/2.2 (ultrawide) + 2MP, f/2.4 (macro), LED flash, panorama, HDR; Front Camera: 13MP, f/2.0, Android 14, up to 6 major Android upgrades, One UI 6.1
  • 3G: HSDPA 850/900/1700(AWS)/1900/2100; 4G LTE: 1/2/3/4/5/7/12/13/14/20/25/26/28/29/30/38/39/40/41/48/66/71, 5G: 2/5/25/41/66/71/77/78 SA/NSA/Sub6/mmWave - Nano-SIM + eSIM
  • US Model – Global Connectivity – Compatible with Most GSM Carriers like T-Mobile, AT&T, MetroPCS, etc. Will Also work with CDMA Carriers Such as Verizon, Straight Talk.

His account describes agents sleeping after 30 seconds of inactivity, along with batching, health checks, automatic restarts, and sparse activation. These are implementation choices he reports using, not guarantees that the same settings will suit another device or workload.

What went wrong, and what the author did about it

The practical problems in the account are familiar constraints for sustained workloads on a phone: memory pressure, network latency, battery draw, process failures, and heat. Hiotis reports two to three agent crashes per day and thermal throttling after about 10 minutes of heavy load. He says he used automatic restarts and active cooling among his mitigations. The source gives no independent crash log, energy measurement, or temperature reading.

Rank #4
Samsung Galaxy A17 5G Smart Phone 128GB, US 1 Yr Manufacturer Warranty Blue | Large AMOLED, High-Res Camera, Durable Design, Super Fast Charging, Expandable Storage, Circle to Search, 2025
  • YOUR CONTENT, SUPER SMOOTH: The ultra-clear 6.7" FHD+ Super AMOLED display of Galaxy A17 5G helps bring your content to life, whether you're scrolling through recipes or video chatting with loved ones.¹
  • LIVE FAST. CHARGE FASTER: Focus more on the moment and less on your battery percentage with Galaxy A17 5G. Super Fast Charging powers up your battery so you can get back to life sooner.²
  • MEMORIES MADE PICTURE PERFECT: Capture every angle in stunning clarity, from wide family photos to close-ups of friends, with the triple-lens camera on Galaxy A17 5G.
  • NEED MORE STORAGE? WE HAVE YOU COVERED: With an improved 2TB of expandable storage, Galaxy A17 5G makes it easy to keep cherished photos, videos and important files readily accessible whenever you need them.³
  • BUILT TO LAST: With an improved IP54 rating, Galaxy A17 5G is even more durable than before.⁴ It’s built to resist splashes and dust and comes with a stronger yet slimmer Gorilla Glass Victus front and Glass Fiber Reinforced Polymer back.

He recommends beginning with three agents and scaling only after routing works. That is the author’s advice, rather than a universally tested threshold. A small first version makes it easier to determine whether task decomposition, message delivery, concurrency control, and synthesis work before adding more roles.

Special offer. See more information about Outbyte and uninstall instructions. Please review EULA and Privacy policy.
Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.Support on Ko-Fi

What the reported results do—and do not—show

Hiotis characterizes Q4_K_M quantization as delivering “90% quality at 30% size,” but his article gives no evaluation method or benchmark for that comparison. It should be read as his claim, not as a general quality-versus-size result for quantized models.

Best Value
Tracfone Motorola Moto G 2025, 64GB, Saphire Blue (Locked to | Prepaid Smartphone- 6.6" 120Hz Display, 5G , 50MP Quad Pixel Camera, Long-Lasting Battery
  • Carrier: This phone is locked to Tracfone, which means this device can only be used on the Tracfone wireless network. Tracfone plan required, activating is easy, just 3 steps.
  • DISPLAY: Immersive viewing on a 6.7-inch super-bright 120Hz display with powerful stereo speakers and Bass Boost for cinematic entertainment.
  • CAMERA SYSTEM: Advanced 50MP Quad Pixel camera captures sharp, detailed photos and videos in any lighting condition
  • PERFORMANCE: Lightning-fast 5G connectivity paired with a powerful processor and RAM Boost for smooth multitasking.
  • BATTERY LIFE: Long-lasting 5000mAh battery with TurboPower charging technology delivers hours of power in minutes.

More broadly, the source is a first-person account. It does not establish that the same speed, memory use, quality, crash rate, or thermal behavior will apply to another Android phone. The handset, operating-system build, model configurations, and reproducibility materials are not specified.

What to take from the design

  • Count defined roles separately from simultaneously running agents; the author reports 34 defined agents and four to six active at a time.
  • Make task decomposition, bounded concurrency, and result synthesis the core of the system rather than treating agent count as the goal.
  • Expect a phone workload to involve trade-offs among local inference, hosted inference, network access, memory, battery use, and heat.
  • Start with a few roles and verify routing and failure handling before scaling, as Hiotis advises.

Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.

Leave a Reply

Your email address will not be published. Required fields are marked *

Special offer. See more information about Outbyte and uninstall instructions. Please review EULA and Privacy policy.

More from the Wire

  1. World desk4 min
    How to Spot an AI Voice Scam Before Sending MoneyDon’t rely on how a caller sounds. Pause, call back through a known number, and verify the emergency with another trusted person before sending money.
  2. Mountain View desk4 min
    Google’s SynthID Detector: How to Check AI-Generated Images, Video and AudioGoogle’s SynthID Detector looks for an embedded watermark in supported images, video and audio. Here is what its results do—and do not—show.
  3. Redmond desk20 min
    How to create a link to File or Folder in Windows 11Windows 11 gives you several ways to point to a file or folder without moving or duplicating it. You can create a desktop shortcut,…
Recommended PC Tool
Recommended PC Tool
PC Slower Than It Used to Be?Free scan - under a minute
Crashes, No Sound, or Screen Glitches?Free driver scan

Two free Windows tools

One Free Minute Could Fix That PC

Before you go - each of these free tools takes about a minute and tackles what quietly slows a Windows PC down.

Special offer. View Outbyte info, uninstall instructions, EULA, and Privacy Policy.