ModelRefs / GPT-5 Mini - AI model implementation reference

GPT-5 Mini - AI model implementation reference

GPT-5 Mini is OpenAI's August 7, 2025 hosted reasoning model for well-defined, latency-sensitive, and high-volume tasks. It accepts text and image input and produces text, supports configurable reasoning effort and verbosity, and has a 400,000-token context window. It is now a previous-generation model, so production decisions should pin gpt-5-mini-2025-08-07 and compare it with current successors.

Overview

GPT-5 Mini is attributed to OpenAI in ModelRefs' canonical registry. Tracked modalities: Text input and output, Image input. Primary use cases considered on ModelRefs: High-volume assistants, extraction, and summarization with measured quality targets; Cost- and latency-sensitive coding and tool-calling workflows.

This ModelRefs profile is decision-support material, not a final or universal ranking. Confirm current behavior, access, pricing, limits, licensing, and lifecycle in OpenAI's own documentation, and evaluate GPT-5 Mini on representative workloads before implementation.

Benchmark & Evaluation

ModelRefs currently has partial, narrow benchmark coverage for GPT-5 Mini. Treat the available benchmark evidence as one input to the decision, not a guarantee that GPT-5 Mini is the strongest option for your workload, and evaluate it on representative workloads before selecting it.

  • Seven canonical provider-run records cover SWE-bench Verified, Aider Polyglot, GPQA Diamond, AIME 2025, tau2-bench telecom, BrowseComp Long Context, and LongFact Concepts for GPT-5 Mini at high reasoning where the source specifies it.
  • These results are source-scoped and provider-reported rather than independent reproduction. OpenAI publishes methodology qualifications and a publication date, but not the physical execution dates; ModelRefs does not infer them or transfer GPT-5 flagship scores.

Implementation considerations

  • Benchmark the mini variant directly against GPT-5 and GPT-5 Nano on representative tasks before committing the routing split.
  • Test structured-output, tool-call, and long-input failure modes at the reasoning-effort level you will run in production.
  • Available as gpt-5-mini and the immutable gpt-5-mini-2025-08-07 snapshot through the Responses API and Chat Completions API; current model documentation labels GPT-5 Mini as a previous model.
  • The documented API price is $0.25 per million input tokens, $0.025 cached input, and $2 per million output tokens; rate limits and retention controls remain account- and endpoint-specific.

Architecture disclosure

  • OpenAI distinguishes the GPT-5 Mini API reasoning model from ChatGPT's routed GPT-5 system and from gpt-5-chat-latest; benchmark claims must not be transferred between those surfaces.
  • The documented API surface supports text input/output and image input, with audio and video unsupported. OpenAI does not publish parameter count, layer design, tokenizer details, training compute, or serving topology.

Prompts and code examples

Version-pinned Responses API request

import OpenAI from "openai";

const client = new OpenAI();
const response = await client.responses.create({
  model: "gpt-5-mini-2025-08-07",
  input: "Extract the implementation risks and return a concise action list.",
  reasoning: { effort: "medium" },
  text: { verbosity: "low" },
});

console.log(response.output_text);

Provider implementation source

Training disclosure

  • OpenAI publishes a May 31, 2024 knowledge cutoff but not a complete training-data inventory, parameter count, compute ledger, or reproducible training recipe.
  • OpenAI states that API customer data is not used to train models unless the customer explicitly opts in; this policy does not describe the original GPT-5 Mini training corpus.
  • The current GPT-5 Mini model page marks fine-tuning as unsupported for this snapshot.

History and source-reviewed changelog

  1. 2025-08-07 — GPT-5 Mini released in the API

    OpenAI released GPT-5 Mini with the GPT-5 API family and documented reasoning effort, verbosity, custom tools, pricing, and variant-specific provider evaluations.

    Milestone source
  2. 2025-08-07 — Immutable GPT-5 Mini snapshot published

    OpenAI documents gpt-5-mini-2025-08-07 as the snapshot that locks behavior and performance for the GPT-5 Mini API model.

    Milestone source
  3. 2026-08-10 — Lifecycle status reviewed

    OpenAI's current model documentation describes GPT-5 Mini as a previous model and recommends a newer generation for new low-latency, high-volume work; ModelRefs retains it for migration and implementation-reference decisions.

    Milestone source

Risks and limitations

  • Hosted-model behavior, quotas, pricing, and data controls can change without a client-side version pin unless a dated snapshot is used.
  • Provider-reported capabilities require task-specific evaluation before production reliance.

Source coverage

ModelRefs verified the GPT-5 Mini launch, current model documentation, system card, implementation controls, API example, data controls, and seven source-scoped provider evaluations. OpenAI does not publish the evaluation execution dates, so ModelRefs preserves them as undisclosed rather than substituting the August 7, 2025 publication date.

Connected ModelRefs evidence

Sources

Continue your research

Use these connected ModelRefs sections to compare alternatives, inspect implementation paths, and review the evidence and governance boundaries relevant to GPT-5 Mini - AI model implementation reference.