GPT-5.4 mini
- Input
- $0.75/M tok
- Output
- $4.50/M tok
- Blended
- $1.688/M tok · 3:1
- Speed
- 217tok/s
- Context
- 400Ktokens
Long-tail import from Artificial Analysis (2026-09-24); AA variant: GPT-5.4 mini (xhigh). Price = AA median host price.
Rankings
Coding
Verified#36 / 103rank by quality45.2±8.0quality / 1002/7 benchmarks measuredWriting & chat
Verified#27 / 146rank by quality69.1±4.2quality / 1003/5 benchmarks measuredVision
Verified#33 / 82rank by quality58.4±6.3quality / 1002/6 benchmarks measuredHard reasoning
Verified#45 / 182rank by quality60.2±3.7quality / 1003/4 benchmarks measuredAgents
Verified#41 / 173rank by quality47.2±4.2quality / 1004/7 benchmarks measured
Benchmark scores
~ italic, dashed = no published score yet, estimated from related benchmarks.
Coding
SWE-bench VerifiedEstimated
~#5 of 14 (estimated)~78.6%
No published score — estimated from related benchmarks
SWE-bench Pro (Public)Estimated
~#13 of 19 (estimated)~57.9%
- #1Claude Opus 5.589.9%
- #2Claude Sonnet 5.581.3%
- #12GPT-5.558.6%
- ~#13GPT-5.4 miniEstimated57.9%
- #13Hy357.9%
No published score — estimated from related benchmarks
DeepSWE v1.1Estimated
~#25 of 31 (estimated)~56.9%
- #1DeepSeek V4.1 Flash74.2%
- #2Grok 4.772.6%
- #24DeepSeek V4 Pro (0813)57.2%
- ~#25GPT-5.4 miniEstimated56.9%
- #25Qwen3.8 2.4T A95B56.6%
No published score — estimated from related benchmarks
Terminal-Bench 4.0Verified
#29 of 832.0%
- #1Claude Sonnet 5.563.6%
- #2Claude Opus 5.559.6%
- #28Gemini 3.1 Pro4.0%
- #29GPT-5.4 mini2.0%
- #29MiniMax-M32.0%
artificialanalysis.ai · 2026-09-24
LiveCodeBenchEstimated
~#6 of 20 (estimated)~80.5%
- #1Qwen3.8-Omni-Flash92.6%
- #2Qwen3.8 Flash-Next91.9%
- #5gpt-oss-120b87.8%
- ~#6GPT-5.4 miniEstimated80.5%
- #6Gemma 4 31B80.0%
No published score — estimated from related benchmarks
SciCodeVerified
#22 of 9552.1%
- #1Claude Opus 5.566.9%
- #2Claude Fable 5.163.1%
- #21GPT-5.5 Instant (June 2026)52.5%
- #22GPT-5.4 mini52.1%
- #22Qwen3.8 Max (0902)52.1%
artificialanalysis.ai · 2026-09-24
AA Coding Agent IndexEstimated
~#15 of 15 (estimated)~41.0
No published score — estimated from related benchmarks
Writing & chat
IFBenchVerified
#21 of 12573.3%
- #1MiniMax-M382.9%
- #2Qwen3.8 2.4T A95B82.8%
- #20Command A+73.9%
- #21GPT-5.4 mini73.3%
- #22Gemma 4 26B A4B72.4%
artificialanalysis.ai · 2026-09-24
LMArena Text, Non-English (Elo)Verified
#23 of 531439 Elo
- #1Claude Opus 5.51501
- #2Claude Fable 5.1Few votes1496
- #22Hy31440
- #23GPT-5.4 mini1439
- #24Gemma 4 31BFew votes1437
lmarena.ai · 2026-09-13
EQ-Bench Creative Writing v3 (Elo)Estimated
~#13 of 26 (estimated)~1835 Elo
- #1GPT-6 Astra2173
- #2Claude Fable 5.12162
- #12Qwen3.8 2.4T A95B1843
- ~#13GPT-5.4 miniEstimated1835
- #13Muse Glimmer1798
No published score — estimated from related benchmarks
Vision
MMMU-ProVerified
#36 of 7673.3%
- #1Claude Opus 5.587.7%
- #2GPT-6 Astra86.9%
- #35Gemma 4 31B73.4%
- #36GPT-5.4 mini73.3%
- #37Qwen3.5 Omni Plus70.5%
artificialanalysis.ai · 2026-09-24
CharXiv ReasoningEstimated
~#6 of 9 (estimated)~78.6%
No published score — estimated from related benchmarks
OmniDocBench v1.5Estimated
~#5 of 8 (estimated)~85.0%
- #1MiniMax-M391.6%
- #2Kimi K391.1%
- #4Gemini 3.1 Pro85.3%
- ~#5GPT-5.4 miniEstimated85.0%
- #5Claude Haiku 4.579.6%
No published score — estimated from related benchmarks
GDP.pdf (AA)Verified
#22 of 4213.4%
- #1GPT-6 Astra31.0%
- #2Muse Spark 1.326.6%
- #21Gemini 3.5 Flash-Lite13.6%
- #22GPT-5.4 mini13.4%
- #23Claude Sonnet 513.2%
artificialanalysis.ai · 2026-09-24
ChartographyEstimated
~#15 of 23 (estimated)~18.9%
- #1GPT-6 Astra71.0%
- #2Claude Opus 5.566.3%
- #14Grok 4.620.5%
- ~#15GPT-5.4 miniEstimated18.9%
- #15Muse Glimmer17.2%
No published score — estimated from related benchmarks
Hard reasoning
GPQA DiamondVerified
#33 of 17287.5%
- #1GPT-6 Astra96.1%
- #2Gemini 3.8 Flash95.3%
- #32Solar Pro 489.1%
- #33GPT-5.4 mini87.5%
- #34K2 Horizon 375B A23B87.3%
artificialanalysis.ai · 2026-09-24
Humanity's Last ExamVerified
#48 of 17828.1%
- #1Claude Opus 5.561.4%
- #2Claude Fable 5.159.1%
- #47GPT-5.4 nano28.3%
- #48GPT-5.4 mini28.1%
- #49MiMo-V2.527.2%
artificialanalysis.ai · 2026-09-24
AIME (latest)Estimated
~#2 of 25 (estimated)~95.2%
No published score — estimated from related benchmarks
Agents
τ²-bench (Telecom)Verified
#24 of 11583.3%
- #1Step 3.7 Flash98.5%
- #2Gemini 3.1 Pro95.6%
- #23GPT-5.3 Codex86.0%
- #24GPT-5.4 mini83.3%
- #24Nemotron 3 Ultra83.3%
artificialanalysis.ai · 2026-09-24
OSWorld-VerifiedEstimated
~#9 of 12 (estimated)~71.2%
- #1Kimi K384.8%
- #2Qwen3.8 27B84.3%
- #8Gemini 3.5 Flash-Lite74.0%
- ~#9GPT-5.4 miniEstimated71.2%
- #9Nex-N2.5-Mini71.2%
No published score — estimated from related benchmarks
OSWorld 2.0Estimated
~last of 5 (estimated)below measured range (<47.5%)
No published score — estimated from related benchmarks
BrowseCompEstimated
~#15 of 18 (estimated)~70.4%
- #1Atria Dawn Preview92.5%
- #2GPT-6 Astra91.5%
- #14Inkling77.1%
- ~#15GPT-5.4 miniEstimated70.4%
- #15Mistral Medium 3.548.6%
No published score — estimated from related benchmarks
GDPval-AA (Elo)Verified
#38 of 861000 Elo
- #1Claude Opus 5.51846
- #2Claude Fable 5.11735
- #37Nemotron 3 Ultra1000
- #38GPT-5.4 mini1000
- #39MiMo-V2.5986
artificialanalysis.ai · 2026-09-24
AutomationBench-AAVerified
#29 of 7926.2%
- #1Claude Opus 5.569.5%
- #2DeepSeek V4.1 Flash68.9%
- #28Apodex 1.128.2%
- #29GPT-5.4 mini26.2%
- #30K2 Horizon MoVA 36B A4B26.1%
artificialanalysis.ai · 2026-09-24