We present HIAPLLM, a privacy-preserving hybrid edge–cloud framework for real-time air-quality advisories tailored to the Indian context using Large Language Models (LLMs). At the network edge, a Raspberry Pi 4 B hosts a Gradio/Folium interface enabling users to input any Indian geocoordinates. Upon each request, the Pi issues a 5 s-timeout call to the Open-Meteo application programming interface (API), retrieving \(\hbox {PM}_{2.5}\) (5.2–82.9 \({\upmu }{\textrm{g}}/{\textrm{m}}^{3}\) ), \(\hbox {PM}_{10}\) (8.6–172.8 \({\upmu }{\textrm{g}}/{\textrm{m}}^{3}\) ), CO (0.10–0.80 \(\hbox {mg/m}^{3}\) ), \(\hbox {NO}_{2}\) (3.6–46.5 \({\upmu }{\textrm{g}}/{\textrm{m}}^{3}\) ), \(\hbox {SO}_{2}\) (0.3–40.7 \({\upmu }{\textrm{g}}/{\textrm{m}}^{3}\) ), \(\hbox {O}_{3}\) (35–160 \({\upmu }{\textrm{g}}/{\textrm{m}}^{3}\) ), aerosol optical depth (AOD: 0.14–1.10), dust loading (0–238 units), and \(\hbox {CH}_{4}\) (1 316–1 614 ppm). These concentrations are mapped via piecewise linear breakpoints into air quality index (AQI) sub-indices and qualitative bands. The Pi then constructs a structured LLM prompt (e.g., “ \(\hbox {PM}_{2.5}\) = 82.9 \({\upmu }{\textrm{g}}/{\textrm{m}}^{3}\) ; Category: Poor”) and forwards it over LAN to a laptop running ten quantized LLMs (0.6–7 B parameters) under the Ollama engine, eliminating external cloud dependency and preserving raw location privacy. We profile model load times (0.80–2.46 s) and inference latencies (6.26–94.05 s), and find a strong inverse relationship between token throughput and total latency (slope − 0.516 s per token/s, \(R^2=0.552\) ). Multivariate analyses (Principal Component Analysis (PCA) capturing 42 %/27 % variance on PC1/PC2; hierarchical clustering) and anomaly detection (Mahalanobis: Kota; Isolation Forest: Delhi, Ludhiana; Density-Based Spatial Clustering of Applications with Noise (DBSCAN) clusters: Hyderabad–Udaipur, Jhansi–Jamshedpur, Belagavi–Nellore) reveal distinct environmental–computational patterns across Indian cities. Variance Inflation Factors ( \(\hbox {PM}_{10}\) : \(8.75\times 10^3\) , dust: \(3.66\times 10^3\) , total_duration_s: \(3.84\times 10^{10}\) ) expose extreme collinearity, addressed via dimension reduction. HIAPLLM’s runtime of \(O(N+P)+T_{\textrm{LLM}}\) and space \(O(N+P+m)\) ensures scalable, transparent, and low-cost deployment suitable for India’s connectivity-constrained, privacy-sensitive settings. Code and data are available at https://github.com/ParthaPRay/Air_Pollution_LLM_Advisor.
