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
- Cost per task
- ~$1.9/taskest.
Long-tail import from Artificial Analysis (2026-09-24); AA variant: GPT-5.4 mini (xhigh). Price = AA median host price.
Rankings
Coding
Verified#63 / 137rank by quality46.6±8.1quality / 1002/7 benchmarks measuredWriting & chat
Verified#37 / 162rank by quality66.5±4.1quality / 1003/5 benchmarks measuredSpeed frontierVision
Verified#45 / 98rank by quality57.4±6.3quality / 1002/6 benchmarks measuredSpeed frontierHard reasoning
Verified#73 / 218rank by quality56.7±3.7quality / 1003/4 benchmarks measuredAgents
Verified#50 / 187rank by quality46.3±4.3quality / 1004/7 benchmarks measured
Benchmark scores
~ italic, dashed = no published score yet, estimated from related benchmarks.
Coding
SWE-bench VerifiedEstimated
~#9 of 21 (estimated)~77.6%
- #1Claude Opus 596.0%
- #2Claude Sonnet 585.2%
- #7Mistral Medium 3.577.6%
- ~#9GPT-5.4 miniEstimated77.6%
- #9Ripple-65B Preview76.5%
No published score — estimated from related benchmarks
SWE-bench Pro (Public)Estimated
~#17 of 24 (estimated)~57.6%
- #1Claude Fable 5.293.0%
- #2Claude Opus 5.589.9%
- #16Hy357.9%
- ~#17GPT-5.4 miniEstimated57.6%
- #17Inkling54.3%
No published score — estimated from related benchmarks
DeepSWE v1.1Estimated
~#37 of 46 (estimated)~54.4%
- #1Gemini 4 Argon78.8%
- #2DeepSeek V4.1 Flash 101777.0%
- #36Nex-N2.5-Pro55.8%
- ~#37GPT-5.4 miniEstimated54.4%
- #37Qwen3.8 Max (0902)51.0%
No published score — estimated from related benchmarks
Terminal-Bench 4.0Verified
#43 of 1012.0%
- #1Kimi K3.1 Pro67.2%
- #2Claude Sonnet 5.563.6%
- #42Gemini 3.1 Pro4.0%
- #43GPT-5.4 mini2.0%
- #43MiniMax-M32.0%
artificialanalysis.ai · 2026-09-24
LiveCodeBenchEstimated
~#15 of 28 (estimated)~73.8%
- #1Qwen3.8-Omni-Flash92.6%
- #2Qwen3.8 Flash-Next91.9%
- #14Maple-Preview75.1%
- ~#15GPT-5.4 miniEstimated73.8%
- #15Mistral Small 464.0%
No published score — estimated from related benchmarks
SciCodeVerified
#38 of 11852.1%
- #1Claude Fable 5.267.7%
- #2Claude Opus 5.566.9%
- #37GPT-5.5 Instant (June 2026)52.5%
- #38GPT-5.4 mini52.1%
- #38Qwen3.8 Max (0902)52.1%
artificialanalysis.ai · 2026-09-24
AA Coding Agent IndexEstimated
~#23 of 26 (estimated)~35.1
- #1Claude Opus 5.561.7
- #2Claude Opus 559.7
- #22Claude Haiku 5.537.3
- ~#23GPT-5.4 miniEstimated35.1
- #23Gemini 3.8 Flash33.3
No published score — estimated from related benchmarks
Writing & chat
LMArena Text, Non-English (Elo)Verified
#25 of 561439 Elo
- #1Gemini 4 ArgonFew votes1512
- #2Claude Opus 5.51497
- #24Hy31440
- #25GPT-5.4 mini1439
- #26Gemma 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
#45 of 8773.3%
- #1Claude Opus 5.587.7%
- #2GPT-6 Astra86.9%
- #44Gemma 4 31B73.4%
- #45GPT-5.4 mini73.3%
- #46Qwen3.5 Omni Plus70.5%
artificialanalysis.ai · 2026-09-24
CharXiv ReasoningEstimated
~#9 of 12 (estimated)~78.7%
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
#27 of 4713.4%
- #1Muse Wonder Full32.6%
- #2GPT-6 Astra31.0%
- #26Gemini 3.5 Flash-Lite13.6%
- #27GPT-5.4 mini13.4%
- #28Claude Sonnet 513.2%
artificialanalysis.ai · 2026-09-24
ChartographyEstimated
~#17 of 25 (estimated)~19.5%
- #1Claude Fable 5.273.4%
- #2GPT-6 Astra71.0%
- #16Grok 4.620.5%
- ~#17GPT-5.4 miniEstimated19.5%
- #17Muse Glimmer17.2%
No published score — estimated from related benchmarks
Hard reasoning
GPQA DiamondVerified
#47 of 19287.5%
- #1GPT-6 Astra96.1%
- #2Gemini 3.8 Flash95.3%
- #46Qwen3.8 27B MXFP487.7%
- #47GPT-5.4 mini87.5%
- #48K2 Horizon 375B A23B87.3%
artificialanalysis.ai · 2026-09-24
Humanity's Last ExamVerified
#63 of 19828.1%
- #1Claude Fable 5.274.3%
- #2Claude Opus 5.561.4%
- #62GPT-5.4 nano28.3%
- #63GPT-5.4 mini28.1%
- #64MiMo-V2.527.2%
artificialanalysis.ai · 2026-09-24
AIME (latest)Estimated
~#8 of 34 (estimated)~93.7%
- #1GLM-5.3 NVFP499.5%
- #2Beam 1.0 501B A23B97.7%
- #7Nova 2.0 Lite94.3%
- ~#8GPT-5.4 miniEstimated93.7%
- #8gpt-oss-120b93.4%
No published score — estimated from related benchmarks
Agents
τ²-bench (Telecom)Verified
#26 of 11983.3%
- #1Step 3.7 Flash98.5%
- #2DeepSeek-V4-Pro-0813-nvfp4-DSpark98.3%
- #25GPT-5.3 Codex86.0%
- #26GPT-5.4 mini83.3%
- #26Nemotron 3 Ultra83.3%
artificialanalysis.ai · 2026-09-24
OSWorld-VerifiedEstimated
~#11 of 15 (estimated)~69.8%
- #1Kimi K384.8%
- #2Qwen3.8 27B84.3%
- #10Nex-N2.5-Mini71.2%
- ~#11GPT-5.4 miniEstimated69.8%
- #11Muse Glimmer65.9%
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
~#18 of 22 (estimated)~71.2%
- #1Atria Dawn Preview92.5%
- #2GPT-6 Astra91.5%
- #17Beam 1.0 501B A23B76.8%
- ~#18GPT-5.4 miniEstimated71.2%
- #18Nemotron 3.5 Bolt 74B-A9B70.0%
No published score — estimated from related benchmarks
AutomationBench-AAVerified
#34 of 8426.2%
- #1Claude Fable 5.276.4%
- #2Muse Wonder Full71.3%
- #33Apodex 1.128.2%
- #34GPT-5.4 mini26.2%
- #35K2 Horizon MoVA 36B A4B26.1%
artificialanalysis.ai · 2026-09-24