ModelsGPT-4o Mini Guide: The Cheap Default and When It Is Enough
GPT-4o mini replaced GPT-3.5 Turbo as the cheap default at $0.15 and $0.60 per million tokens. Where a small model holds up, where it does not, and its 2027 deadline.
ReadNotes on models, agents, and what inference costs
Practical writing for developers building with large language models — how they work, how to pick one, and how to keep the bill predictable.
ModelsModelsGPT-5 Image Mini: Cheap Drafts and Bulk Image Jobs
GPT-5 Image Mini pairs GPT-5 Mini with GPT Image 1 Mini for lower-cost image output. What it costs per picture, what you give up, and a draft-then-finish workflow.
ReadModelsGPT-5 Image: Text Reasoning and Image Output in One Call
GPT-5 Image pairs GPT-5 with image generation behind one model ID. What that means for prompting, billing and quality, and when a dedicated image model is better.
Read
ModelsGPT-5 in 2026: Should You Keep the August 2025 Model?
GPT-5 shipped in August 2025 and OpenAI now lists it as a previous model. What it is, what it still costs, what it introduced, and when migrating off it pays.
ReadModelsGPT-5 Mini Guide: The 2025 Middle Tier, Priced for 2026
GPT-5 Mini is a reasoning model at $0.25 in and $2 out per million tokens. What the 400K window really allows, how effort sets the bill, and when it beats Nano.
Read
ModelsGPT-5 Nano Guide: Reasoning at Five Cents Per Million Tokens
GPT-5 Nano is the smallest GPT-5 model, built for classification and summaries. A worked cost example, where it breaks, and when to step up to GPT-5 Mini.
Read
ModelsGPT-5 Pro Guide: When $120 Per Million Output Tokens Pays Off
GPT-5 Pro spends more compute per answer, runs only at high effort and only on the Responses API. How to call it without timeouts and when the price is justified.
Read
ModelsGPT-5.1 Guide: The Release That Made Reasoning Optional
GPT-5.1 shipped with reasoning switched off by default. What that changed for latency and cost, the specs, the batch endpoint, and the 2027 shutdown date to plan for.
Read
ModelsGPT-5.1-Codex-Max: Compaction, Long Runs and Its Retirement
OpenAI trained Codex-Max to work across several context windows in one task. What compaction changed, what the benchmarks showed, and how to move off it now.
ReadModelsGPT-5.1-Codex-Mini: The Cheap Tier of OpenAI's Coding Line
Codex-Mini cost a fifth of Codex-Max with the same 400K window. Which coding jobs a small model handles, when to escalate, and what replaced it after retirement.
ReadModelsGPT-5.1-Codex: What It Was and Why Listings Still Show It
GPT-5.1-Codex was OpenAI's agentic coding model for the Responses API until its July 2026 retirement. What it did, what it cost, and how to move an agent off it.
ReadModelsGPT-5.2 Chat: The ChatGPT Snapshot and Its Moving Alias
GPT-5.2 Chat cost the same as GPT-5.2 but had a smaller window, a 16K output cap and an alias that changed under you. When it fitted, and why pinning beat it.
ReadShowing 97–108 of 731 articles