BenchLeader

GPT-4o mini

Reasoning effort
Quality#315 of 333
35.4Index
Bottom quarter
−14.6 on the median
Price#52 of 249
$0.262per 1M tokens
Cheapest quarter
0.3× the median
Speed#147 of 215
59tokens/s
Below the median
0.8× the median

45 cheaper ranked models score higher: Claude Haiku 5.5 (57.3, $0.200), GLM 5.3 Flash (56.9, $0.238), GPT-6 Luna (56.7, $0.200) and 42 more.

As of 8 Nov 2026, GPT-4o mini by OpenAI scores 35.4 on the BenchLeader Index (593 ranked configurations), with its best results in human preference (47) and weakest in knowledge (22). API pricing: $0.150 per 1M input tokens and $0.600 per 1M output tokens ($0.262 blended at 3:1), list price on 8 Nov 2026. It streams about 59 tokens/s (Artificial Analysis). Released 18 Jul 2024.

Key facts (as of 8 Nov 2026)

  • GPT-4o mini API pricing: $0.150 per 1M input tokens and $0.600 per 1M output tokens ($0.262 blended at 3:1) (list price, 8 Nov 2026).
  • GPT-4o mini streams about 59 tokens/s (Artificial Analysis, 8 Nov 2026).
  • GPT-4o mini was released on 18 Jul 2024 (OpenAI).

In detail. GPT-4o mini is an OpenAI proprietary model, released 18 Jul 2024. Its best configuration ranks #573 of 593 on the BenchLeader Index at 35.4 ±4.1, in the lower half. The ± is the point: 84 other configurations score within that range, so they and this one cannot be told apart on quality alone — price and speed are what separate them. It scores highest in human preference (47) and lowest in knowledge (22). At $0.262 per million tokens blended it is cheaper than most ranked models. Output speed of 59 tokens per second puts it slower than most, with a first answer in 1.7 s. It has been measured at 2 reasoning-effort settings; this summary describes the best-scoring one, and the tabs above switch between them. Last measured 6 Nov 2026.

Blended price
$0.262/M
$0.150 in · $0.600 out
via Artificial Analysis
Output speed
59 tok/s
measured by Artificial Analysis
First answer
1.68 s
first token 0.58 s
Context
128k
via models.dev
Full answer
31 s
median, reasoning included
Cost per task
–
Tokens per task
–
Released
18 Jul 2024
How it scores by categoryDashed line = average model (50). One step of 15 = one standard deviation.
  1. Overall index35
  2. Reasoning30
  3. Coding29
  4. Agents & tools30
  5. Maths37
  6. Knowledge22
  7. Instruction following42
  8. Human preference47
  9. Multimodal31

Get an email when OpenAI ships a new model

And once when it gets ranked, plus the weekly review on Mondays.

Choose providers and emails

Reasoning-effort configurations

The same model behaves differently depending on how much it is allowed to think. Each row is one setting, scored only on the benchmarks that were run at that setting. “Not stated” collects results from publishers that did not say which setting they used; for a reasoning model that is usually its thinking mode, but we do not assume it. Pick a setting here or at the top of the page to see its price, speed and category scores.

EffortIndexRankSpeedFirst answerChat reply costCategories
high33.5#58259 tok/s1.68 s$0.0002Coding 15 · Knowledge 29 · Maths 35
not statedbest35.4#57359 tok/s1.68 s$0.0002Agents & tools 30 · Coding 29 · Human preference 47 · Instruction following 42 · Knowledge 22 · Maths 37 · Multimodal 31 · Reasoning 30

How its index has moved

8 Sep 2026 to 8 Nov 2026
283441
  • not stated
  • high

The index is recomputed from scratch every day, so a line moves when a new benchmark result lands, when a publisher revises a score, or when the models it is normalised against change. Early movement usually means the score is still settling.

Benchmark results

See all LLM benchmarks. One column per reasoning effort. Rank is among every configuration of every model on that benchmark. Hover a score for the run it came from.

Reasoning

Benchmarkhighnot statedSourceTrend
GPQA Diamondeffort not stated–37.7%#265Epoch AI Benchmarking Hub
SimpleBencheffort not stated–10.7%#89SimpleBench
Arena Hard Promptseffort not stated–1311#260Arena (LMArena)
GPQA Diamond (AA)not in indexeffort not stated–42.6%#473Artificial Analysis
Humanity's Last Exam (AA)not in indexeffort not stated–4.2%#541Artificial Analysis
GPQA Diamond (Vals)not in index44.2%#126–Vals AI
Kagi LLM Benchmarkeffort not stated–28.8%#126Kagi LLM Benchmark
GPQA Diamond (HELM)not in indexeffort not stated–36.8%#53HELM Capabilities
ARC-AGI-2effort not stated–0.0%#245ARC Prize
Chess Puzzlesnot in indexeffort not stated–3.0%#146Epoch AI Benchmarking Hub
Mystery Game Puzzlesnot in indexeffort not stated–12.0%#106Epoch AI Benchmarking Hub
LMCAnot in indexeffort not stated–10.4%#180Epoch AI Benchmarking Hub
DTBenchnot in indexeffort not stated–54.4%#201Epoch AI Benchmarking Hub
PIQAnot in indexeffort not stated–88.7%#1Epoch AI Benchmarking Hub

Coding

Benchmarkhighnot statedSourceTrend
WeirdMLeffort not stated–11.8%#151WeirdML
Arena Codingeffort not stated–1348#254Arena (LMArena)
LiveCodeBench26.4%#137–Vals AI
Aider Polygloteffort not stated–3.6%#49Aider polyglot leaderboard
Terminal-Bench 2.1 (AA)not in indexeffort not stated–5.6%#207Artificial Analysis
LiveCodeBench (AA)not in indexeffort not stated–23.4%#232Artificial Analysis

Agents & tools

Benchmarknot statedSourceTrend
τ³-Banking (AA)2.9%#203Artificial Analysis
BALROGnot in index17.4%#27BALROG
BrowseCompnot in index0.6%#25OpenAI

Maths

Benchmarkhighnot statedSourceTrend
FrontierMath Tiers 1–3effort not stated–0.7%#118Epoch AI Benchmarking Hub
OTIS Mock AIMEeffort not stated–6.9%#249Epoch AI Benchmarking Hub
MATH Level 5effort not stated–52.6%#53Epoch AI Benchmarking Hub
AIME (Vals)11.5%#86–Vals AI
MGSM86.2%#62–Vals AI
Omni-MATH (HELM)effort not stated–28.0%#48HELM Capabilities
MATH 500not in index72.6%#47–Vals AI
Arena Mathseffort not stated–1275#257Arena (LMArena)
GSM8Knot in indexeffort not stated–91.3%#5Epoch AI Benchmarking Hub

Knowledge

Benchmarkhighnot statedSourceTrend
SimpleQA Verifiedeffort not stated–8.3%#96Epoch AI Benchmarking Hub
MMLU-Pro62.7%#130–Vals AI
CorpFin45.5%#114–Vals AI
TaxEval60.5%#125–Vals AI
MedQA72.4%#79–Vals AI
MMLU-Pro (HELM)not in indexeffort not stated–60.3%#47HELM Capabilities
MMLUnot in indexeffort not stated–81.8%#16Epoch AI Benchmarking Hub

Instruction following

Human preference

Benchmarknot statedSourceTrend
Arena Text1318#246Arena (LMArena)
Arena Creative Writingnot in index1294#229Arena (LMArena)
Arena Multi-turnnot in index1311#244Arena (LMArena)
Arena Longer Queriesnot in index1326#238Arena (LMArena)
Lech Mazur Creative Writingnot in index67.2%#36Lech Mazur

Multimodal

Benchmarkhighnot statedSourceTrend
Arena Visioneffort not stated–1066#132Arena (LMArena)
MMMU-Proeffort not stated–41.5%#276Artificial Analysis
MMMU (validation)effort not stated–59.4%#34MMMU
MMMU-Pro (official)not in indexeffort not stated–37.6%#24MMMU
MMMU-Pro (Vals)not in index56.6%#87–Vals AI
MortgageTaxnot in index54.5%#77–Vals AI
GeoBenchnot in indexeffort not stated–64.0%#14GeoBench
VPCTnot in indexeffort not stated–34.0%#27Epoch AI Benchmarking Hub
Video-MMEnot in indexeffort not stated–64.8%#17Video-MME

Composite

Safety & honesty

Benchmarknot statedSourceTrend
FORTRESSnot in index48.1%#17Scale AI SEAL

API pricing: where to run it

Every provider serving this model through OpenRouter, with throughput and first-token latency measured on live traffic over the last 30 minutes and each provider’s own price. Purple marks the best in each column.

ProviderSpeedFirst tokenInput $/MOutput $/MContextQuantisation
Azure87 tok/s0.75 s$0.165$0.660128k–
Azure59 tok/s1.19 s$0.150$0.600128k–
OpenAI56 tok/s0.59 s$0.150$0.600128k–

What a task costs

Estimates from list price, output speed and time to first answer for the best configuration. “With caching” assumes three-quarters of the input is served from the prompt cache at $0.075 per 1M. Reasoning tokens are not modelled.

WorkloadTokens in / outCostWith cachingTime
Chat reply400 / 300$0.0002$0.00026.8 s
Summarize a 30-page report12,000 / 600$0.0022$0.001511.8 s
Code edit6,000 / 1,500$0.0018$0.001527.1 s
Agentic coding session60,000 / 4,000$0.011$0.00801.2 min
Structured extraction2,000 / 200$0.0004$0.00035.1 s

More from OpenAI

New this month

See also

Embed or cite

Show where GPT-4o mini ranks on a model card or README. The badge redraws itself with every daily update, carries the data date and links back to this page. To see ranks like this on OpenRouter and Hugging Face model pages, get the Chrome extension.

[![GPT-4o mini on BenchLeader: #307 overall](https://www.benchleader.com/badge/gpt-4o-mini.svg)](https://www.benchleader.com/models/gpt-4o-mini)

Cite this

The ranking changes daily, so a reference carries the data date. APA style:

BenchLeader. (2026, November 8). GPT-4o mini on the BenchLeader Index (method v1.4) [Data set]. Retrieved November 8, 2026, from https://www.benchleader.com/models/gpt-4o-mini

BibTeX:

@misc{benchleader_gpt_4o_mini_20261108,
  author       = {{BenchLeader}},
  title        = {{GPT-4o mini on the BenchLeader Index (method v1.4)}},
  year         = {2026},
  month        = nov,
  howpublished = {\url{https://www.benchleader.com/models/gpt-4o-mini}},
  note         = {Data as of 2026-11-08}
}

Every badge style, the embeddable top-10 widget and how to cite the Index

Questions people ask

What are GPT-4o mini's benchmark scores?
As of 8 Nov 2026, GPT-4o mini scores 35.4 ±4.1 on the BenchLeader Index, strongest in human preference (47) and weakest in knowledge (22). That rests on independent benchmark results only, and every score on this page links to its publisher.
How does GPT-4o mini rank?
GPT-4o mini is #573 of 593 ranked configurations on 8 Nov 2026. 84 other configurations score within its margin, so the exact position matters less than the band. The leader is Claude Opus 5.5 (max effort) at 69.4.
How much does GPT-4o mini cost per 1M tokens?
As of 8 Nov 2026, GPT-4o mini costs $0.150 per 1M input tokens and $0.600 per 1M output tokens, $0.262 blended (three input tokens per output token).
When was GPT-4o mini released?
OpenAI released GPT-4o mini on 18 Jul 2024. BenchLeader has results for it at 2 reasoning-effort settings.

Data as of 8 Nov 2026. Compare these configurations.

Cite as: BenchLeader, “GPT-4o mini: benchmarks, pricing, speed and rank”, data as of 8 Nov 2026, https://www.benchleader.com/models/gpt-4o-mini.