GPT-5.4 MINI
128K context. Vision input. Native web search. $0.75/$4.50 per million tokens — roughly 70% cheaper than GPT-5.4 and a great balance of capability and cost for everyday tasks.
GPT-5.4 Mini is OpenAI's budget-friendly variant in the GPT-5.4 series — the model that replaces GPT-5 Mini for everyday chat, classification, drafting, and light reasoning. It ships with a 128K context window, vision input, native web search via the Responses API, and the same tool-calling shape as its bigger siblings. Priced at $0.75 per million input tokens and $4.50 per million output, it's roughly 70% cheaper than GPT-5.4 ($2.50/$15) and one-seventh the cost of GPT-5.5 ($5/$30). The case for GPT-5.4 Mini: high-volume tasks where you don't want to burn flagship dollars — chatbots, summarization, classification, customer support flows, light coding. The case against: frontier reasoning (use GPT-5.5 or 5.4), refactor-heavy coding (Sonnet 4.6 or Opus 4.7), and 200K+ token jobs (it tops out at 128K).
Council AI exposes GPT-5.4 Mini on every paid plan inside multi-model councils. You don't manage API keys, rate limits, or per-call pricing — the model is bundled into your monthly plan budget, and the orchestrator picks it (or routes around it) based on the task. When the council runs, GPT-5.4 Mini produces its independent answer in parallel with the other selected models, then the Moderator synthesizes a single response that reconciles the disagreements.
GPT-5.4 Mini is the smaller, cheaper sibling — 128K context vs 400K, ~70% cheaper on input, and tuned for everyday tasks rather than frontier reasoning.
$0.75 per million input tokens and $4.50 per million output tokens. Council AI bundles it into monthly plan budgets so high-volume use stays predictable.
Yes to both — vision input is native, and web search runs through the OpenAI Responses API automatically when Council AI routes the call.
Yes — it's arguably the best fit in the OpenAI lineup for production chatbots. Cheap enough for high volume, smart enough for nuanced replies, and stable on tool calls.
Use Mini when quality matters and you want vision and web search. Use Nano for the cheapest, fastest path on extremely high-volume background tasks.
GPT-5.4 Mini — same role in the lineup, upgraded core, and the GPT-5.4 family's improved tool calling.