PulseAugur
EN
LIVE 02:30:12

28.9M-parameter LLM runs on $8 microcontroller using Google's Per-Layer Embeddings · 4 sources tracked

By PulseAugur Editorial · [4 sources] ·

A developer has successfully run a 28.9 million parameter language model on an $8 ESP32-S3 microcontroller, achieving approximately 9 tokens per second without cloud dependency. This significant advancement in edge AI leverages Google's Per-Layer Embeddings technique, allowing most of the model's parameters to reside in slow flash memory while keeping the core processing components in the chip's limited SRAM. The model, trained on the TinyStories dataset, generates short, coherent narratives, demonstrating a new capability for low-cost, offline generative AI applications. AI

IMPACT Enables sophisticated AI capabilities on extremely low-cost, power-efficient edge devices, opening new possibilities for offline smart sensors and embedded systems.

RANK_REASON Project demonstrates running a large LLM on commodity hardware, which is a significant tooling advancement for edge AI.

Read on Mastodon — fosstodon.org →

AI-generated summary · Google Gemini · from 4 sources. How we write summaries →

28.9M-parameter LLM runs on $8 microcontroller using Google's Per-Layer Embeddings · 4 sources tracked

How we ranked this

Signal score
0 / 100
Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket
Tool
Project demonstrates running a large LLM on commodity hardware, which is a significant tooling advancement for edge AI.
Source corroboration
4 independent sources
Strong cross-source corroboration — multiple independent publishers covered this within the clustering window.
Topics
product, infra
Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance
High
Clearly on-topic for AI-industry coverage.
Story freshness
46 days old
Aged out of breaking-news scoring windows; ranking reflects the durable signal from the full source set.

Full methodology in our editorial standards.

COVERAGE [4]

  1. Hacker News — AI stories ≥50 points TIER_1 English(EN) · boveyking ·

    Running a 28.9M parameter LLM on an $8 microcontroller

  2. dev.to — LLM tag TIER_1 English(EN) · Hamza ·

    ESP32-AI: Running a 28.9M-Parameter LLM on an $8 Microcontroller

    <p><a class="article-body-image-wrapper" href="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto%2Cformat=auto/https%3A%2F%2Fdev-to-uploads.s3.us-east-2.amazonaws.com%2Fuploads%2Farticles%2F0zoirw9smd3py4jw342s.jpg"><img alt="ESP32-S3 microco…

  3. Mastodon — fosstodon.org TIER_1 English(EN) · [email protected] ·

    Running a 28.9M parameter LLM on an $8 microcontroller https:// github.com/slvDev/esp32-ai # ai # github # llm

    Running a 28.9M parameter LLM on an $8 microcontroller https:// github.com/slvDev/esp32-ai # ai # github # llm

    LINKS github.com/…/esp32-ai
  4. Mastodon — mastodon.social TIER_1 English(EN) · h4ckernews ·

    Running a 28.9M parameter LLM on an $8 microcontroller https:// github.com/slvDev/esp32-ai Comments: https:// news.ycombinator.com/item?id=4 9050512 # HackerNew

    Running a 28.9M parameter LLM on an $8 microcontroller https:// github.com/slvDev/esp32-ai Comments: https:// news.ycombinator.com/item?id=4 9050512 # HackerNews # LLM # microcontroller # AI # lowcost # techinnovation

    LINKS github.com/…/esp32-ai

RELATED ENTITIES

RELATED TOPICS