Brainscope/examples/ESP32 Watch a microcontroller's LLM think
Watch a microcontroller's LLM think This model normally lives on an ESP32-S3 microcontroller - a $5 chip with 512 KB of RAM - where it writes children's stories at 9.88 tokens/s on a matchbox-sized board (slvDev/esp32-ai). It fits because 25M of its 28.9M parameters sit in the chip's flash memory (Per-Layer Embeddings, the Gemma 3n trick) and are read ~450 bytes per token. Here you get those exact weights - the int4 artifact the chip runs, dequantized and verified against its C runtime to ~1e-5 - under brainscope's microscope.
- ▪Watch a microcontroller's LLM think This model normally lives on an ESP32-S3 microcontroller - a $5 chip with 512 KB of RAM - where it writes children's stories at 9.88 tokens/s on a matchbox-sized board (slvDev/esp32-ai).
- ▪It fits because 25M of its 28.9M parameters sit in the chip's flash memory (Per-Layer Embeddings, the Gemma 3n trick) and are read ~450 bytes per token.
- ▪Here you get those exact weights - the int4 artifact the chip runs, dequantized and verified against its C runtime to ~1e-5 - under brainscope's microscope.
Hacker News (AI / LLM) files mainly under ai. We currently carry 3,896 of its stories.
Story provenance
Source · retrieval · rights · ranking — open for full record
inspect →
Story provenance
Attribution is not the same as permission. This drawer separates discovery metadata, excerpts, WeSearch-generated summaries, reuse status, and whether the publisher receives the visit. Nothing here claims a legal grant the publisher has not made.
Record
| Original publisher | GitHub |
| Canonical URL | https://github.com/moudrkat/brainscope/tree/main/examples/esp32 |
| Publication time | Thu, 06 Aug 2026 21:10:42 +0000 |
| Retrieval time | 2026-08-06T21:15:47.872Z |
| Last seen | 2026-08-06T21:15:47.872Z |
| Headline source | Publisher (no WeSearch rewrite) |
| Excerpt source | publisher body |
| Excerpt method | First ~120 words (~800 chars) of extracted publisher body, fair-use limited. |
| Summary | WeSearch · cerebras-chat (WeSearch summarizer) |
| Summary source text | contentText |
| Citation coverage | Summary is a WeSearch-generated derivative; primary citation is the original publisher URL. |
| Cluster | 069pq-CJGFWb · 1 stories |
| Cluster logic | Grouped by semantic title/content similarity across sources within a rolling window. Same-publisher template collisions are excluded from coverage comparison. |
| Ranking reason | Story pages are not engagement-ranked. Hub feeds use recency, with optional source-diversified chronological ordering (cap consecutive stories per source). No personalized ranking. |
| Publisher visit | Yes — open original |
| Substitutes article? | No — link-out required for full text |
Rights status (four layers)
WeSearch handling by dimension
| Indexing | May the item be indexed (stored, ranked, made findable)? | Allowed |
| Snippet | May a short excerpt of the publisher's text be shown? | Allowed |
| AI summary | May WeSearch generate its own short summary of the article? | Limited |
| Retrieval / RAG | May the content be exposed for third-party retrieval-augmented generation? | Not asserted |
| Model training | May the content be used to train AI models? | Not asserted |
| Commercial reuse | May the content be reused commercially? | Not permitted |
Basis: Derived from the published RSS/Atom feed. Contact: [email protected]. Reviewed: 2026-07-24.
Opening excerpt (first ~120 words) tap to expand
Watch a microcontroller's LLM think This model normally lives on an ESP32-S3 microcontroller - a $5 chip with 512 KB of RAM - where it writes children's stories at 9.88 tokens/s on a matchbox-sized board (slvDev/esp32-ai). It fits because 25M of its 28.9M parameters sit in the chip's flash memory (Per-Layer Embeddings, the Gemma 3n trick) and are read ~450 bytes per token. Here you get those exact weights - the int4 artifact the chip runs, dequantized and verified against its C runtime to ~1e-5 - under brainscope's microscope. Six layers, four heads: the whole model fits on one screen. No cherry-picked attention heads, no truncated views. It is the perfect glass-box model for learning what the logit lens, attention maps and steering actually show.
…
Excerpt limited to ~120 words for fair-use compliance. The full article is at GitHub.