Choosing a model
-
Fable vs Opus: When the Premium Model Is Worth It
A practical walkthrough for testing Fable against Opus, measuring total cost, and deciding when a premium model earns its place.
-
A Simple Rubric for Comparing Model Answers
Compare model answers with a small, repeatable rubric that checks correctness, usefulness, and cost before you spend more.
-
Compare Context Windows Without Wasting Tokens
A step-by-step way to compare model context limits, test useful recall, and choose enough room without paying for unnecessary tokens.
-
Choose a Model for Translating Documentation
A practical walkthrough for testing translation quality, preserving technical details, and keeping documentation costs predictable.
-
Pick the Right Model for Your Hackathon Weekend
A practical walkthrough for testing models, setting a spending limit, and shipping a hackathon demo without burning your budget.
-
GLM and MiniMax: Two Options Worth Testing
A budget-conscious walkthrough for testing GLM-5.3 and MiniMax M3 on real tasks, comparing results, and deciding when to use them.
-
A Practical Qwen Workflow for Multilingual Writing
Build a repeatable Qwen writing workflow with clear briefs, small tests, locale checks, and request-level cost tracking.
-
Grok for Brainstorming: A Practical Walkthrough
Use Grok to explore ideas, challenge assumptions, and choose a small experiment without letting brainstorming consume your budget.
-
DeepSeek V4 Reasoning on a Small Budget
A step-by-step walkthrough for testing DeepSeek V4, controlling token use, and checking reasoning results without stretching your budget.
-
Which Model Should Summarise Your Lecture Notes?
A practical walkthrough for choosing a model, testing summary quality and keeping lecture-note summarisation costs under control.
-
Gemini Flash for Fast Drafts and Quick Answers
A step-by-step walkthrough for using Gemini Flash to draft, ask focused questions, and manage your AI spending with MyRelay.
-
Haiku for Cheap Background Tasks: A Practical Walkthrough
Build a budget-conscious background workflow with Claude Haiku 4.5, clear task boundaries, validation, and request-level cost tracking.
-
GPT-6.1 Sol vs GPT-6 Astra: Pick One for Coding
A budget-conscious walkthrough for comparing GPT-6.1 Sol and GPT-6 Astra on real coding tasks before choosing a daily model.
-
Claude Opus vs Sonnet for a Solo Developer
A step-by-step workflow for choosing Sonnet or Opus, checking results, and keeping everyday AI development costs under control.