I'm making LLMs work on small and old devices
The README of the smolmaxxing GitHub repo describes running LLMs of varying sizes (260K to 16B parameters) on low-resource devices, including a Xiaomi TV Box S with about 1 GB usable RAM and a 2013 iPad Air with 1 GB RAM.
Source-based brief · Editor review not recorded · See checks and limits ↓
One primary source; other links are context, not independent corroboration.
- Primary sourceGitHub
- Discovery linkHacker News
A community post points to the original link. Discussion and upvotes do not establish accuracy.
Open full article view →- Community discussion+5
- Fetched source page+30
- No editor link review+0
- Community link+0
This measures the stored source trail and editorial steps. It does not certify accuracy, peer review, or independent corroboration. Popularity adds no points.
View full scoring →Read the scoring rubric →What the source says
The README of the smolmaxxing GitHub repo describes running LLMs of varying sizes (260K to 16B parameters) on low-resource devices, including a Xiaomi TV Box S with about 1 GB usable RAM and a 2013 iPad Air with 1 GB RAM. It reports measured token rates, such as Qwen3.5-0.8B at 16.8 tok/s and Qwen3.5-2B at 7.7 tok/s on the TV box, and a 1.7B model at 24-26 tok/s on the iPad. Techniques mentioned include custom kernels, memory management, and weight streaming from flash. The full article was not checked beyond the supplied README excerpt.
Claim explorer
Attribution, source and conditionsSource-stated points from the stored brief. Claim-level assessments have not been recorded.
Source claims not independently checked; figures are self-reported by the repo author and measured on specific devices with variable thermal and background conditions.
This connects a source, an attributed statement and its qualifications. It does not establish that the claim is correct.
Questions worth asking
Open questions for further reportingWhat remains unconfirmed?
Source claims not independently checked; figures are self-reported by the repo author and measured on specific devices with variable thermal and background conditions.
What material was available for this brief?
The recorded brief basis is fetched source page. These points are AINewsIQ paraphrases, not quotations or independent confirmations.
Reports Qwen3.5-0.8B at 16.8 tok/s and Qwen3.5-2B at 7.7 tok/s on a Xiaomi TV Box S with about 1 GB usable RAM.
Source: GitHub ↗Stored source-based brief · key point 1 · AINewsIQ paraphraseReports a 1.7B model at 24-26 tok/s on a first-generation iPad Air (2013) CPU-only, jailbroken.
Source: GitHub ↗Stored source-based brief · key point 2 · AINewsIQ paraphraseDescribes an 8B mixture-of-experts model whose experts are paged from flash on demand, at 7.2 tok/s on the TV box.
Source: GitHub ↗Stored source-based brief · key point 3 · AINewsIQ paraphraseLimits from the stored brief
Source claims not independently checked; figures are self-reported by the repo author and measured on specific devices with variable thermal and background conditions.
This is a source-based abstract, not independent validation of the source's claims.
Evidence & sources
These links show where the signal came from and what the source page linked to. A linked page is context, not independent corroboration.
Related archive coverage
These relationships record a citation or topic overlap, not confirmation of the finding.
A community post points to the original link. Discussion and upvotes do not establish accuracy.
Timeline & updates
- Signal dated by source or discovery feed
This timestamp places the signal in the radar. It may be the discovery post date rather than the primary page's first publication date.
- Shared brief stored
AINewsIQ prepared one abstract from fetched source page.
Brief history · 1 retained version
These are changes to AINewsIQ’s own publication. Earlier history is unavailable unless recorded here. Complete retained versions are available to the editor.
- Version 1 · published
Source-based brief stored or updated. This does not establish that the publisher changed its page.
Automated record
Source observations · 1 recent captures
A changed capture means the extracted text differed. It can reflect page furniture or extraction changes; it does not establish a substantive correction. Source captures remain private.
Retained capture; no earlier comparison available.
New engagement snapshots do not mean the source itself changed. Source observations and AINewsIQ brief revisions are separate records.
Why this may matter
Shows concrete measured throughput for LLM inference on very old, memory-constrained consumer hardware, which informs what edge deployment is practical today.
Limits and open questions
Source claims not independently checked; figures are self-reported by the repo author and measured on specific devices with variable thermal and background conditions.
Organization-specific financial, security, and operational effects require the organization's own context and review.

Join the discussion
Have a different perspective or a question about the evidence? Discussion is free with a verified email account. Every comment is reviewed before publication.
No published comments yet.
New commenters: 2 submissions per 24 hours, at least 5 minutes apart, until 3 are approved. Discussion guidelines and moderation →