Needle-rs – AI Function calling in the browser, 258 KB WASM
A new AI tool called needle-rs has been developed, allowing function calling directly in the browser using WebAssembly. This tool operates entirely on the user's device without the need for a server or API key. The model features a 26M-parameter transformer and is designed for efficient performance with a runtime of 258 KB.
- ▪Needle-rs is a working AI agent implemented in 258 KB of WebAssembly.
- ▪The model operates without requiring a server or API key, ensuring data privacy.
- ▪It features a 26M-parameter tool-calling transformer with a call latency of approximately 280 ms.
Hacker News (AI / LLM) files mainly under ai. We currently carry 3,301 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 | Pages |
| Canonical URL | https://needle-rs.pages.dev |
| Publication time | Mon, 18 May 2026 20:35:33 +0000 |
| Retrieval time | 2026-05-18T20:39:56.949Z |
| Last seen | 2026-05-18T20:39:56.949Z |
| 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 | D4t_CCUsL8yo |
| 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
needle-rs AI TOOL CALLING · WASM · NO_STD GitHub A working AI agent in 258 KB of WebAssembly. Below is a 26M-parameter tool-calling transformer running entirely in this tab — no server, no API key, no data leaving your device. The model is Needle by Cactus Compute; needle-rs is the pure-Rust runtime that makes it deployable here. 258 KB Runtime 22 MB Weights ~280 ms Per call
Excerpt limited to ~120 words for fair-use compliance. The full article is at Pages.
