LLMs consume 5.4x less mobile energy than ad-supported web search
Details
- External ID
- 47899803
- Source
- HN
- Company
- —
- Product
- LLMs consume 5.4x less mobile energy than ad-supported web search
- Website domain
- dupr.at
- Launched
- April 25, 2026
- Cohort
- —
- Upvotes
- 19
- Upvotes percentile
- 0.7410025706940874
- Tags
- —
- Fetched at
- Sept. 7, 2026, 9:26 p.m.
- Updated at
- Sept. 7, 2026, 9:26 p.m.
Description
The standard AI energy debate compares server-side LLM inference to a server-side Google query. I think this misses most of what actually happens on a mobile device during a real search session.I built a parametric model of the full end-to-end mobile search session: 4G/5G radio energy, SoC rendering cost for a 2.5MB page, programmatic advertising RTB auctions running in the background, and network transmission costs for both sides. Then compared it to an equivalent LLM session.Main finding across 10,000 Monte Carlo draws: on mobile, a standard LLM session uses on average 5.4x less energy than a classic ad-supported web search session. Programmatic advertising alone accounts for up to 41% of device battery drain per session.Caveats I tried to be explicit about:- Advantage disappears on fixed Wi-Fi/fiber- Reverses for reasoning models- Parametric model, not empirical device measurement. Greenspector has offered to run terminal measurements for v2- Jevons paradox appliesSSRN working paper, not peer-reviewed. Methodology and Monte Carlo distributions fully documented in the paper. Happy to defend the assumptions.DOI: 10.2139/ssrn.6287918
Enrichment
- Theme
- ML inference and model optimization
- Vertical
- Horizontal
- Function
- Analytics & BI
- Audience
- B2B
- AI stance
- AI feature
- Project type
- Commercial product
- Normalized one-liner
- energy efficiency analysis of llm-based search
- Manually corrected
- False
Could you build this?
No This is novel hardware and thermodynamics research involving empirical mobile radio (4G/5G) state profiling, SoC power measurement, and thermodynamic mathematical modeling.
What it would actually take: Producing this research and simulation requires hardware power measurement equipment (e.g., Monsoon power monitors) connected to real mobile testbeds to benchmark baseband radio power states (RRC state machines) and mobile GPU/CPU rendering loads. Developing the parametric thermodynamic model requires graduate-level expertise in mobile computing systems, telecommunications standards, and device-level energy modeling.
Discussion
1 comment analyzed.
Concerns raised: chats will catch up with ads in effectiveness
Competitors
Other products that read as similar to this one — 57 launches clear the similarity bar, closest 8 shown.
Attention rank: #10 of 58 (itself plus its competitors, highest first — normalized so YC and Product Hunt are compared fairly).
Launched 172 days after the earliest competitor.
- Running PrismML's Bonsai inside DRAM by breaking DDR4 timing rules · hn · 2026-07-23 · 23 upvotes · similarity 0.41
- LLM Inference Calculator · hn · 2026-08-28 · 6 upvotes · similarity 0.40
- Token Economics Calculator for AI inference hardware · hn · 2025-11-19 · 13 upvotes · similarity 0.40
- IronMule · ph · 2026-09-17 · 1 upvotes · similarity 0.38
- Frugon · hn · 2026-07-07 · 67 upvotes · similarity 0.38
- ClawRouter · hn · 2026-02-05 · 12 upvotes · similarity 0.38
- Needle2: 14MB agentic LLM for phones, wearables, smart home and robots · hn · 2026-08-10 · 537 upvotes · similarity 0.37
- Llm.sql · hn · 2026-04-24 · 8 upvotes · similarity 0.37
Other launches for this product
- No other launches for this product.
Same idea, different domain
Nobody's really built a analytics & bi tool for Legal yet.