Models/GPT-5.1 Codex mini

MDL

GPT-5.1 Codex mini

OpenAI

Verified against Artificial Analysis · Jul 13, 2026

Capability

percentile index

Price /M

$0.25 / $2

in / out

Context

400K

tokens

Avg score

62.5

10 benchmarks

AA Intelligence

30.6

index

Speed

204 tok/s

TTFT 4.8s

Released

Nov 13, 2025

About

GPT-5.1 Codex mini is a proprietary model in the GPT family from OpenAI. Benchmarked on 10 evals, averaging 62.5.

Benchmark scores · 10

avg 62.5 — higher is better

AIMEartificial-analysis · 2026-07

91.7

LiveCodeBenchartificial-analysis · 2026-07

83.6

MMLU-Proartificial-analysis · 2026-07

GPQA Diamondartificial-analysis · 2026-07

81.3

IFBenchartificial-analysis · 2026-07

67.9

τ²-benchartificial-analysis · 2026-07

62.9

AA Long-Context Reasoningartificial-analysis · 2026-07

62.7

SciCodeartificial-analysis · 2026-07

42.6

Terminal-Bench Hardartificial-analysis · 2026-07

33.3

Humanity's Last Examartificial-analysis · 2026-07

16.9

Details

PublisherOpenAI

FamilyGPT

Parameters—

License—

WeightsClosed

ReleasedNov 13, 2025

Performance & price

Output speed204 tok/s

Latency (TTFT)4.8s

Price in / out$0.25 / $2 per 1M

Blended price$0.688 / 1M

Math index91.7

Median across API providers, via Artificial Analysis. Blended = 3:1 input:output. For reasoning models, latency includes thinking time. Methodology →

History

full ledger →

Capability over time

Append-only ledger — every observed change to this model's numbers.

API providers

Artificial Analysis$2/M