GPT-4o-mini

GPT-4o mini carries forward the architecture and design philosophy of its larger GPT-4o sibling, but in a distilled form that prioritizes cost efficiency and broad accessibility. The model was engineered to make powerful AI available to developers and businesses at a fraction of the price, scoring 82% on the MMLU benchmark and outperforming its larger cousin on chat preferences in the LMSYS leaderboard. Its improved tokenizer, shared with GPT-4o, makes handling non-English text particularly cost effective, expanding its utility across global markets and diverse language applications.

The lineage traces clearly to GPT-4o through a distillation process that compresses the larger model's capabilities into a leaner, more economical package without sacrificing core performance. This approach positions GPT-4o mini as an ideal foundation for high-volume, latency-sensitive workflows—handling tasks like chaining multiple API calls, processing lengthy code bases or conversation histories, and powering real-time customer support chatbots. While vision capabilities are available today, the roadmap includes support for video and audio modalities, suggesting this model is built not just for present needs but for an increasingly multimodal AI landscape.

Poeopenai/gpt-4o-minigpt-mini

Open provider page Provider docs

Quick Info

Provider: Poe
Model key: openai/gpt-4o-mini
Release date: Jul 18, 2024
Last updated: Jul 18, 2024
Input modalities
Output modalities
Capabilities

Cost

Input token cost: $0.14
Output token cost: $0.54

Limits

Output tokens: 4,096 tokens
Context window: 124,096 tokens

Transparent token rates

Compare GPT-4o-mini pricing

Rates are shown per one million tokens. Combined means one million input plus one million output tokens.

Browse this family

Latest news about GPT-4o-mini

Poe

CoverageBenchmark

Magica's September 2026 spec page for GPT-4o-mini describes the model as developed by OpenAI with a 128K-token input context window, a 16.4K output token limit, an October 31, 2023 knowledge cutoff, and a release date of July 18, 2024. Pricing is listed at $0.15 per million input tokens, $0.60 per million output tokens The page reports a Coding Index score of 11.4 from Artificial Analysis for GPT-4o-mini and positions the model within a broader comparison grid against other contemporary models such as GLM 5.3 Flash, DeepSeek V4.1 Flash, GPT-5.6 Luna, and DeepSeek V4 Flash variants across categories including programming, creative wri

Poe

CoverageBenchmark

Artificial Analysis' page for GPT-4o mini flags the model as deprecated, stating "We only continue performance benchmarking for the default 10k input token workload" and recommending OpenAI's newer GPT-5 mini (medium) as a replacement. The page reports an Artificial Analysis Intelligence Index score of 7 out of 80 (med Technical specifications on the page list GPT-4o mini as a non-reasoning model supporting text and image input with text output, a 128K-token context window, and an October 1, 2023 knowledge cutoff. The summary characterizes the model as below average in intelligence but well priced compared with other non-reasoning mo