{"slug": "gemma-4-analyzed-my-bank-statements-apparently-i-have-a-problem-with-coffee-and", "title": "\"Gemma 4 Analyzed My Bank Statements – Apparently I 'Have a Problem' with Coffee and Late-Night Apps\"", "summary": "The article describes a project submission for the Gemma 4 Challenge, where the author built a \"Bank Statement Analyzer\" that uses the Gemma 4 26B A4B instruction-tuned model. The tool allows users to upload 3–6 months of bank statements to receive a breakdown of spending patterns, forgotten subscriptions, anomalies, and cost-cutting suggestions. The author explains that Gemma 4's long context window, structured extraction capabilities, and Mixture-of-Experts efficiency were ideal for parsing the statements and generating actionable insights.", "body_md": "This is a submission for the Gemma 4 Challenge: Build with Gemma 4\nWhat I Built\nBank statement Analyzer — upload 3–6 months of statements, get a breakdown of spending patterns, subscriptions you forgot about, anomalies, and concrete suggestions to cut your costs.\nDemo\nCode\nAbelCodeCanvas / my-bank-app\nBank statement Analyzer — upload 3–6 months of statements, get a breakdown of spending patterns, subscriptions you forgot about, anomalies, and concrete suggestions to cut your costs.\nmarkdown\n💰 Bank Statement Analyser\nUpload 3–6 months of bank statements and get a clear breakdown of:\n- 📊 Spending patterns – where your money really goes\n- 🔁 Subscriptions you forgot about – recurring charges you might not need\n-\n⚠️ Anomalies – unusual or unexpected transactions - ✂️ Concrete suggestions – actionable advice to cut costs\nPowered by Gemma 4 26B A4B instruction‑tuned model via Hugging Face.\n📋 Prerequisites\nBefore you begin, make sure your local machine has:\n- Python 3.9 or higher (recommended: 3.10)\n- Git – to clone the repository\n- A Hugging Face account (free) with a User Access Token Create one here\n- At least 16 GB RAM (32 GB recommended)\n- GPU with 12+ GB VRAM (optional but strongly recommended for fast inference) – if no GPU, the app will fall back to CPU (very slow for 26B model)\nNote: The 26B A4B model is large but uses Mixture‑of‑Experts to reduce compute…\nHow I Used Gemma 4\nFor my Bank Statement Analyser, I used Gemma 4 26B A4B (the instruction-tuned variant) on Hugging Face. While not exactly one of the standard sizes (E2B, E4B, or 31B Dense), this 26B parameter model strikes an ideal balance for the task:\nLong context handling – Bank statements over 3–6 months contain hundreds of transactions. The model’s large context window lets me feed entire statements without chunking, preserving temporal patterns.\nStructured extraction – Gemma 4’s instruction-tuning excels at parsing semi-structured data (PDF/CSV statements) and outputting consistent JSON breakdowns of spending, subscriptions, and anomalies.\nReasoning for suggestions – The 26B size provides enough reasoning capacity to identify cost-cutting opportunities (e.g., duplicate subscriptions, high-fee accounts, irregular charges) without the latency or cost of a dense 31B model.\nA4B efficiency – The Mixture-of-Experts (A4B) architecture reduces compute per token, making it feasible to run locally or on a free Hugging Face T4 GPU.\nIn short, Gemma 4 powers the entire pipeline: statement parsing → spending categorization → anomaly detection → actionable recommendations.", "url": "https://wpnews.pro/news/gemma-4-analyzed-my-bank-statements-apparently-i-have-a-problem-with-coffee-and", "canonical_source": "https://dev.to/abelmhlanga/gemma-4-analyzed-my-bank-statements-apparently-i-have-a-problem-with-coffee-and-late-night-pkf", "published_at": "2026-05-22 12:27:10+00:00", "updated_at": "2026-05-22 12:34:58.253005+00:00", "lang": "en", "topics": ["artificial-intelligence", "machine-learning", "large-language-models", "open-source", "developer-tools"], "entities": ["Gemma 4", "Hugging Face", "AbelCodeCanvas", "Bank Statement Analyzer"], "alternates": {"html": "https://wpnews.pro/news/gemma-4-analyzed-my-bank-statements-apparently-i-have-a-problem-with-coffee-and", "markdown": "https://wpnews.pro/news/gemma-4-analyzed-my-bank-statements-apparently-i-have-a-problem-with-coffee-and.md", "text": "https://wpnews.pro/news/gemma-4-analyzed-my-bank-statements-apparently-i-have-a-problem-with-coffee-and.txt", "jsonld": "https://wpnews.pro/news/gemma-4-analyzed-my-bank-statements-apparently-i-have-a-problem-with-coffee-and.jsonld"}}