GPT-5 Mini (2025-08-07)
Aug 2025
GPT-5 Mini (2025-08-07)Compare inon
Benchmarks of GPT-5 Mini (2025-08-07)27
BenchmarkByScoreRank
Knowledge2
Reasoning1
Math5
Writing2
Legacy · saturated or no longer reported17
No benchmark matches
Rankingat high effort?
Alternatives?
ModelScoreCostPer taskWhat changes
CheaperGPT-5.4 Nano (2026-03-17)high-5%-33%$0.20 / $1.25
BetterClaude Opus 4.8max+73%+1,355%$4.08
OpenKimi K2.6unstated+43%+149%$0.95 / $4Context: 400K → 256K
CheaperQwen 3.5 Plusunstated-23%-62%$0.12 / $0.69BetterGPT-5.4 Pro (2026-03-05)+84%+9,718%$30 / $180No strict JSONModelScoreCostPer taskWhat changes
GPT-5.4 Nano (2026-03-17)high-5%-33%$0.20 / $1.25Qwen 3.5 Plusunstated-23%-62%$0.12 / $0.69ModelScoreCostPer taskWhat changes
Claude Opus 4.8max+73%+1,355%$4.08GPT-5.4 Pro (2026-03-05)+84%+9,718%$30 / $180No strict JSONGPT-5.4 (2026-03-05)xhigh+75%+718%$2.50 / $15Claude Opus 4.7xhigh+61%+1,355%$5 / $25Claude Opus 4.6max+49%+1,355%$5 / $25ModelScoreCostPer taskWhat changes
Kimi K2.6unstatedOpen+43%+149%$0.95 / $4Context: 400K → 256K
GLM 5.1unstatedOpen+23%+213%$1.40 / $4.40Context: 400K → 203K+1Kimi K2 ThinkingunstatedOpen-21%+24%$0.47 / $2Context: 400K → 256K+1GLM 5unstatedOpen-40%+125%$1 / $3.20Context: 400K → 203K+1Qwen3 235B A22B Thinking 2507unstatedOpen-69%+9%$0.23 / $2.30Context: 400K → 262K+2ModelScoreCostPer taskWhat changes
GPT-5.4 Pro (2026-03-05)+84%—$30 / $180No strict JSONGPT-5.4 (2026-03-05)xhigh+75%—$2.50 / $15GPT-5.2 (2025-12-11)high+48%—$1.75 / $14GPT-5 (2025-08-07)high+19%—$1.25 / $10GPT-5.1 (2025-11-13)high+14%—$1.25 / $10GPT-5.4 Mini (2026-03-17)high+4%—$0.75 / $4.50GPT-5.4 Nano (2026-03-17)high-5%—$0.20 / $1.25o3 (2025-04-16)high-31%—$2 / $8Context: 400K → 200KAll OpenAI models →History
7 versionsBenchmarks
all categoriesChat & Knowledge
SimpleQA Verified v121.6#67
MATH Level 597.8#2
FrontierMath27.2#15
FrontierMath Tier 46.3#14
FrontierMath Tiers 1–3 (v2)46.7#47
FrontierMath Tier 4 (v2)12.2#47
Chess Puzzles24#44
GPQA Diamond · Epoch75#75
OTIS Mock AIME 2024 202586.7#55
Coding
SWE-rebench · all · 2025-12-01 to 2026-02-0139.6#2
SWE-rebench · Go · 2025-12-01 to 2026-02-010#1
SWE-rebench · Java · 2025-12-01 to 2026-02-010#1
SWE-rebench · Rust · 2025-12-01 to 2026-02-010#1
SWE-rebench · Python · 2025-12-01 to 2026-02-0139.6#2
SWE-rebench · TypeScript · 2025-12-01 to 2026-02-010#1
FAQ
How much does GPT-5 Mini (2025-08-07) cost?›
$0.25 per 1M input tokens and $2 per 1M output tokens.
Is GPT-5 Mini (2025-08-07) free to use?›
There is no free API; the cheapest paid price is $0.25 per 1M input tokens.
How good is GPT-5 Mini (2025-08-07), and what is it best at?›
#67 of 83 in Chat & Knowledge, #19 of 24 in Coding, #8 of 39 in Agents & Tool Use, #58 of 95 in Writing & Language.
Is GPT-5 Mini (2025-08-07) better than GPT-5.4 Mini (2026-03-17)?›
On FrontierMath, GPT-5 Mini (2025-08-07) scores 27.2 vs 28.3 — lower.
How can I use GPT-5 Mini (2025-08-07)?›
Through the OpenAI API.
What is the context window of GPT-5 Mini (2025-08-07)?›
400K tokens.
When was GPT-5 Mini (2025-08-07) released? Is there a newer version?›
Released 2025-08-07. It has been superseded by GPT-5.6 Terra.
Can GPT-5 Mini (2025-08-07) read images?›
It accepts text and images as input.
What is GPT-5 Mini (2025-08-07)?›
GPT-5 Mini (2025-08-07) is a language model from OpenAI, released Aug 2025.
What are the best alternatives to GPT-5 Mini (2025-08-07)?›
GPT-5.4 Nano (2026-03-17) (cheaper), Claude Opus 4.8 (better), Kimi K2.6 (open weights).