modelbenchmark.io

GPT-4.1 mini

OpenAI · gpt-4-1-mini

Compare

Affordable GPT-4.1 lane for fast coding help and structured extraction

Specification

openai listing

Context
1M
Max output
33K
Released
2025-04-14
Knowledge cutoff
2024-04
Retires
2027-04-14
Open weights
no
Input
text, image, pdf
Output
text

Price

US dollars per million tokens · openai list price

Input
$0.40
Output
$1.60
Cache read
$0.10
Cache write
$0.40

Batch and priority prices

US dollars per million tokens · per listing

PriceListing$/M
Batch inputazure$0.20
Batch inputazure$0.22
Batch outputazure$0.80
Batch outputazure$0.88
Priority inputopenai$0.70
Priority outputopenai$2.80

Batch input is 50% below the interactive input price at the same listing.

Quality

10 benchmarks · 3 sources

Composite
31st
percentile of 334 scored models
Rank
230 / 334
±0.05 sd
Evidence
10 × 3
3 or more benchmarks, from 2 or more sources
Effort range
one configuration only
coding16th2/3 bench
reasoning39th3/4 bench
math40th4/7 bench
knowledge9th1/1 bench

Every score

one row per source and configuration — nothing averaged away

BenchmarkScoreConfigurationSourceRun
MATH level 587.3 ±0.7defaultEpoch AI2025-04-14
GPQA diamond65.8 ±2.7defaultEpoch AI2025-04-14
OTIS Mock AIME 2024-202544.7 ±6.1defaultEpoch AI2025-04-14
Chess Puzzles7.0 ±2.6defaultEpoch AI2026-07-15
Aider polyglot32.4default · aider diffAider2025-04-14
FrontierMath-v17.2 ±10.0default · 2 runsEpoch AI2025-04-14
Mystery Game Puzzles7.0 ±2.6defaultEpoch AI2026-08-27
FrontierMath-Tiers-1-3-v26.7 ±1.5defaultEpoch AI2026-08-27
SimpleQA Verified12.7 ±1.1defaultEpoch AI2026-08-31
SWE-Bench verified23.9default · mini-SWE-agentSWE-bench

The leading figure for a benchmark is the median across its configurations, so one heroic high-effort run cannot set the number.

Available from 28 hosts

HostIn $/MOut $/MCache rdCache wrContextOutputRetires
openaivendor$0.40$1.60$0.101M33K
poe$0.36$1.40$0.091M33K
302ai$0.40$1.601M33K
abacus$0.40$1.60$0.101M33K
aixy$0.40$1.60$0.101M33K
azure
$0.40
$0.44
$1.60
$1.76
$0.10
$0.11
1M33K2027-04-14
azure-cognitive-services$0.40$1.60$0.101M33K
cloudflare-ai-gateway$0.40$1.60$0.101M33K
edenai$0.40$1.60$0.101M33K
helicone$0.40$1.60$0.101M33K
impossibl$0.40$1.60$0.101M33K
kilo$0.40$1.60$0.101M33K
llmgateway$0.40$1.60$0.101M33K
llmgateway-providers$0.40$1.60$0.101M33K
merge-gateway$0.40$1.60$0.101M33K
nano-gpt$0.40$1.60$0.101M33K
nearai$0.40$1.60$0.101M33K
ofox$0.40$1.60$0.101M33K
openrouter$0.40$1.60$0.101M33K
orcarouter$0.40$1.60$0.101M33K
pioneer$0.40$1.60$0.20$0.401M33K
replicate · replicate$0.40$1.60
sap-ai-core$0.40$1.60$0.101M33K
vercel$0.40$1.60$0.101M33K
vercel_ai_gateway · vercel-ai-gateway$0.40$1.60$0.10free1M33K
cortecs$0.434$1.704$0.1341M33K
anyapi1M33K
model-oracle-ai1M33K

More from OpenAI

most-hosted first