gpt-5.6-luna
Input price$0.2
Output price$1.2
OpenAITextVision
gpt-4o-mini
gpt-4o-mini is OpenAI's model at $0.15 input and $0.60 output per million tokens, with a 128K context window. On a typical 3:1 input-to-output mix that blends to roughly $0.26 per million tokens, placing it in the budget band of the 89 models tracked here — cheaper than 75% of them.
Context window128K
Input price$0.15 / 1M
Output price$0.6 / 1M
Output speed160 tok/s
Quality score80
Data updatedUpdated Jun 2026
Price verified2026-08-19 · Official
By price, gpt-4o-mini ranks 19th cheapest of 89 on input and 22nd on output. Its quality score of 80 ranks 61st of 89, and on value (quality score ÷ output price) it ranks 20th of 89. Its blended rate sits 91% below the catalog average.
In real billing terms: a support chatbot handling 50,000 turns a month (2,000 input / 500 output tokens each) costs about $30 per month on gpt-4o-mini, while a RAG workload of 20,000 queries (8,000 input / 800 output tokens each) runs about $34. Because output is priced 4.0× input here, long-generation tasks scale cost fastest — cap output length before you optimise anything else.
A 128K context window holds roughly 171 pages of A4 text (at ~750 tokens per page), ranking 74th of 89 here. That covers typical RAG and multi-turn chat, though book-length inputs still need retrieval and chunking. Output runs at about 160 tok/s (21st fastest of 89), fast enough for real-time interaction and high-throughput batches.
We track 15 OpenAI models, and gpt-4o-mini's blended rate sits 94% below that provider's average. If your task tolerates a little less headroom, qwen3.7-flash (Alibaba) blends about 79% cheaper at a quality score of 82, and is worth benchmarking first. Going the other way, mercury-2.5-preview scores 82 on quality for about 74% less.
The closest model in OpenAI's line-up is gpt-6-luna ($0.10 input / $0.50 output, quality —). Against it, gpt-4o-mini scores 80 points higher on quality, has a 922K smaller context window, costs 24% more blended. When two models sit this close on price, the deciding factor is usually the quality ceiling — run both against your real prompts before picking.
Our read on it
Still running in production everywhere, and its price advantage is now much smaller than people assume.
The default nobody re-evaluated
gpt-4o-mini is $0.15/$0.60, quality 80, 128K context, 160 tokens/second. Through 2024 and 2025 it more or less defined the budget API category — it was the model in the tutorials, the SDK samples and the starter templates — so a great deal of production traffic still runs on it today. Usually not because somebody compared it recently, but because nobody touched it.
In absolute terms it is still cheap: roughly $0.26 per 1M tokens blended. What has changed is that being cheap this way is no longer distinctive.
Three live alternatives, side by side
Its own sibling gpt-5.6-luna costs $0.20/$1.20 — more per token, but ten points higher on quality, with 1,049K of context (8x) and audio support. Alibaba's qwen3.5 is $0.10/$0.15, cheaper than 4o-mini on both sides, and scores 91. InclusionAI's Ling-3.0-flash is $0.07/$0.22, also cheaper, matching the quality score at 80 with double the context at 262K.
So 4o-mini is now squeezed from both directions at once: cheaper-and-better below it, slightly-dearer-and-much-better above it. The 128K context is the most visible weakness — it is on the small side for its own price band in 2026.
When keeping it is still the right call
Two defensible reasons. Ecosystem: OpenAI's SDK, structured outputs, function calling and the surrounding tooling remain the most mature, and switching is not free engineering. Stability: a system that has run for a year with prompts tuned against this specific model needs a full regression pass before the model underneath it changes, and that cost is real.
For a new project, though, there is no case for making 4o-mini the default. The same prompt is usually cheaper and more accurate on qwen3.5, and gets eight times the context plus audio on luna. The useful move is unglamorous: run your existing eval set against all three and compare pass rate against bill. It takes less than a day and the answer is rarely ambiguous.
Estimated monthly cost
Monthly API spend across four common workloads, next to the cheapest model we track.
| Workload | Usage assumption | This model | Cheapest model |
|---|---|---|---|
| Support chatbot | 2,000 input / 500 output tokens per turn, 50,000 turns per month | $30 | $3.00 |
| RAG document Q&A | 8,000 input / 800 output tokens per query, 20,000 queries per month | $34 | $3.84 |
| Coding assistant | 12,000 input / 2,000 output tokens per request, 10,000 requests per month | $30 | $3.20 |
| Batch summarization | 4,000 input / 400 output tokens per item, 200,000 items per month | $168 | $19 |
Derived directly from the rates on this page. Excludes prompt caching, batch discounts and free tiers.
How it compares
The alternatives worth benchmarking alongside it, from the same dataset.
| Model | Provider | Blended /1M | Quality | Price delta |
|---|---|---|---|---|
| gpt-4o-mini | OpenAI | $0.26 | 80 | — |
| qwen3.7-flashCheaper option | Alibaba | $0.055 | 82 | -79% |
| mercury-2.5-previewHigher quality | Inception | $0.068 | 82 | -74% |
| gpt-6-lunaSame provider | OpenAI | $0.20 | — | -24% |
Best for
ChatbotAffordableFast
Modalities
TextVision
Where it fits — and where it doesn't
Good fit
- Chat and support: natural multi-turn conversation for user-facing products.
- High throughput and real-time UX: 160 tok/s keeps streamed responses feeling instant.
- High-volume, cost-sensitive batch work: low rates suit classification, tagging and summarization at scale.
- Multimodal input: handles images, screenshots and scanned documents directly.
Poor fit
- Hard reasoning: a quality score of 80 is on the low side here; pick a stronger model for complex planning or maths.
Frequently asked questions
How much does the gpt-4o-mini API cost?
gpt-4o-mini costs $0.15 per million input tokens and $0.60 per million output tokens. On a 3:1 input-to-output mix that blends to about $0.26 per million tokens.
What does gpt-4o-mini cost per month?
It depends on volume. Using the four worked scenarios on this page: a support chatbot runs about $30/month, RAG document Q&A about $34/month, a coding assistant about $30/month, and batch summarization about $168/month. Plug your own token counts into the cost calculator for a tailored figure.
How large is the gpt-4o-mini context window?
128K tokens — roughly 171 pages of A4 text at ~750 tokens per page, ranking 74th of the 89 models tracked here.
gpt-4o-mini or qwen3.7-flash — which is better value?
qwen3.7-flash blends to about $0.055 per million tokens versus $0.26 for gpt-4o-mini — roughly 79% cheaper — at a quality score of 82 against 80. If that quality gap does not show up on your task, take the cheaper one; if it does, stay with gpt-4o-mini. Benchmark both on your own data before deciding.
What is gpt-4o-mini best suited for?
Tagged use cases: chatbot, cheap, fast. Chat and support: natural multi-turn conversation for user-facing products. Less suitable: Hard reasoning: a quality score of 80 is on the low side here; pick a stronger model for complex planning or maths.
How current is this gpt-4o-mini pricing?
Pricing here was last verified on 2026-08-19 against official documentation. llmprice.app runs a daily collection job, but providers can change rates between runs — confirm against OpenAI's official pricing page before you commit.
Estimate the real cost of this model
Bring tokens per request and monthly volume into the calculator to compare total model costs.
Related models
Continue by provider and use case.
View model
gpt-5.4-nano
Input price$0.2
Output price$1.25
gpt-5.4-mini
Input price$0.75
Output price$4.5
Pricing, speed and quality scores change over time. Confirm official documentation before production use.