13
real model files check the formula
71
guides, all sourced and dated
Will that model fit on your GPU?
Pick a model to see how much memory it needs and which NVIDIA GeForce cards have enough.
Check one in ten seconds
Pick a model and its quantization.
7.3 GBEstimated memory with 8K tokens of context
Memory sizes, filled where it fits
Fits in 8 GB
19 GeForce cards have enough: RTX 5090, RTX 5080, RTX 5070 Ti, RTX 5070 and 15 more.
Latest guides
Newest first.
LLM APIs & CostsPrices & updates
How much a chatbot costs per month: three traffic levels, eight models
What a support chatbot costs in API fees on eight models at 1,000, 10,000 and 100,000 conversations a month, and how much prompt caching saves.
Language Models ExplainedExplainer
What a system prompt is and how to write one
A system prompt sets a model's role, rules and format before the user types. How OpenAI, Claude and Gemini handle it, what it costs and why it is not secret.
Rules & SecurityGuide
Open-weight model licenses: what you can build with Gemma, Qwen, Muse, Llama and more
What open-weight model licenses allow: Apache 2.0 and MIT models, Qwen, Muse and Llama custom licenses, revenue thresholds, attribution and the EU clause.
Language Models ExplainedExplainer
What is a mixture of experts (MoE) model?
Mixture of experts explained: how MoE models route each token to a few experts, why total parameters set memory and active ones set speed, and how to run them.
Coding AssistantsComparison
Automated code review: Copilot, Claude Code and Codex compared
Automated code review compared: GitHub Copilot code review, Claude Code /code-review and Codex, how each one works, what it costs and how to get good results.
Agents & AutomationHow-to
How to self-host n8n: Docker, editions and the Sustainable Use License
How to self-host n8n with Docker: requirements, the recommended Compose setup, what the free Community edition leaves out and what the license allows.
Topics
Pick an area to see all its guides.
Assistants & Chatbots
16 guides
ChatGPT, Claude, Gemini and Copilot: plans, limits and prices compared.
LLM APIs & Costs
11 guides
Token prices, context windows and caching, with a cost calculator.
Agents & Automation
8 guides
What agents are, which business tasks suit them and what they cost.
Local Models & Hardware
11 guides
How much memory a model needs, which graphics card fits it and how to run it.
Language Models Explained
14 guides
Tokens, context, RAG, quantization and hallucinations, explained without hype.
Coding Assistants
6 guides
Coding assistants compared by price, models and workflow.
Rules & Security
5 guides
EU Regulation 2024/1689 and other rules: what they require and from when.
Three kinds of guide
Pick by what you need: compare, do or understand.
Comparisons
Plans, prices and limits side by side, each with its source.
How-to guides
One job, step by step, with what it costs to run.
Explainers
What a thing is, how it works and when it matters.
What changed recently
What we added or fixed most recently.
| Date | Guide | What changed | Topic |
|---|---|---|---|
| How much a chatbot costs per month: three traffic levels, eight models | First published. Checked against the pricing and caching documentation of OpenAI, Anthropic, Google and DeepSeek on October 28, 2026. | LLM APIs & Costs | |
| What a system prompt is and how to write one | First published. Checked against OpenAI, Anthropic and Google documentation and the OWASP entry on system prompt leakage on October 28, 2026. | Language Models Explained | |
| Open-weight model licenses: what you can build with Gemma, Qwen, Muse, Llama and more | First published. Checked on October 23, 2026 against the license files and model cards of Gemma 4, gpt-oss, Qwen3.8, DeepSeek V4.1, Muse Spark Open, Muse Glimmer and Llama 4, Mistral's models overview and the Apache License 2.0. | Rules & Security | |
| What is a mixture of experts (MoE) model? | First published. Checked on October 23, 2026 against the Shazeer et al. 2017, Switch Transformers and Mixtral papers, Hugging Face's Mixture of Experts Explained, the gpt-oss README, the Qwen3.8-2.4T-A95B model card, Meta's Muse Spark Open announcement and the llama.cpp server README. | Language Models Explained | |
| Automated code review: Copilot, Claude Code and Codex compared | First published. Checked on October 23, 2026 against GitHub's Copilot code review documentation, Anthropic's /code-review and ultrareview documentation and OpenAI's Codex code review and GitHub integration documentation. | Coding Assistants |
How we work
Every figure has a source and the date we checked it.
No pay-to-rank
Vendors cannot pay to be listed or ranked. We say what we have and have not tested.
374
sources linked from our guides
- Primary sources
- Vendor pricing pages, official docs, model cards and the law itself.
- Dated
- Each guide shows when we checked it and when it is due for review.
- guides published
- 71
- topics
- 7
- editions: English and Spanish
- 2
- last update
- Nov 6, 2026