GPT-5.4 mini
- Context
- 400K
- 128K max output
- Input price
- $0.75
- per 1M tokens · OpenAI API
- Output price
- $4.50
- per 1M tokens
- Epoch Capabilities Index
- 148.8
- #46 of 71 listed
- LMArena Text
- 1,447
- #44 of 79 listed
Where to run it
3 offerings · prices as listed| Meter | Tier | Price |
|---|---|---|
| Input | Standard | $0.75 / 1M tokens |
| Batch | $0.38 / 1M tokens | |
| Flex tier | $0.38 / 1M tokens | |
| Fast tier | $1.50 / 1M tokens | |
| Output | Standard | $4.50 / 1M tokens |
| Batch | $2.25 / 1M tokens | |
| Flex tier | $2.25 / 1M tokens | |
| Fast tier | $9.00 / 1M tokens | |
| Cache read | Standard | $0.075 / 1M tokens |
| Batch | $0.038 / 1M tokens | |
| Flex tier | $0.038 / 1M tokens | |
| Fast tier | $0.15 / 1M tokens | |
Azure OpenAI in Foundry (opens in a new tab) gpt-5-4-mini | ||
| Input | Standard | $0.75 / 1M tokens |
| Batch | $0.38 / 1M tokens | |
| Priority tier | $1.50 / 1M tokens | |
| Output | Standard | $4.50 / 1M tokens |
| Batch | $2.25 / 1M tokens | |
| Priority tier | $9.00 / 1M tokens | |
| Cache read | Standard | $0.080 / 1M tokens |
| Batch | $0.040 / 1M tokens | |
| Priority tier | $0.15 / 1M tokens | |
Azure OpenAI in Foundry (opens in a new tab) gpt-5-4-mini | ||
| Input | Standard | $0.83 / 1M tokens |
| Priority tier | $1.65 / 1M tokens | |
| Output | Standard | $4.95 / 1M tokens |
| Priority tier | $9.90 / 1M tokens | |
| Cache read | Standard | $0.090 / 1M tokens |
| Priority tier | $0.17 / 1M tokens | |
Cited scores
From third-party leaderboards, not measured by us- Epoch Capabilities Index 148.8
#46 of 71 listed models · CI 146.9–150.4
Epoch AI, 'Capabilities & benchmarking'. Published online at epoch.ai. Retrieved from 'https://epoch.ai/benchmarks'. Licensed CC BY 4.0. · License · as of Sep 26, 2026 · changes we made
- LMArena Text 1,447
#44 of 79 listed models · high effort · CI 1,443–1,451 · 63,140 votes
LMArena (Arena), leaderboard-dataset on Hugging Face, licensed CC BY 4.0 · License · as of Sep 25, 2026 · changes we made
- Epoch AI · FrontierMath Tiers 1–3 51.2%
#37 of 56 listed models · xhigh effort · ±2.97
Also listed: low effort 24.6% · reasoning off 17.2%
Epoch AI, 'Capabilities & benchmarking'. Published online at epoch.ai. Retrieved from 'https://epoch.ai/benchmarks'. Licensed CC BY 4.0. · License · as of Sep 26, 2026 · changes we made
- Epoch AI · GPQA Diamond 86.9%
#41 of 64 listed models · xhigh effort · ±2.41
Also listed: high effort 83.6% · reasoning off 64.1%
Epoch AI, 'Capabilities & benchmarking'. Published online at epoch.ai. Retrieved from 'https://epoch.ai/benchmarks'. Licensed CC BY 4.0. · License · as of Sep 26, 2026 · changes we made
- LMArena Text · Chinese 1,474
#53 of 79 listed models · high effort · CI 1,464–1,484 · 3,931 votes
LMArena (Arena), leaderboard-dataset on Hugging Face, licensed CC BY 4.0 · License · as of Sep 25, 2026 · changes we made
- LMArena Text · Coding 1,495
#46 of 79 listed models · high effort · CI 1,489–1,501 · 17,234 votes
LMArena (Arena), leaderboard-dataset on Hugging Face, licensed CC BY 4.0 · License · as of Sep 25, 2026 · changes we made
- LMArena Vision 1,245
#37 of 49 listed models · high effort · CI 1,239–1,252 · 24,041 votes
LMArena (Arena), leaderboard-dataset on Hugging Face, licensed CC BY 4.0 · License · as of Sep 13, 2026 · changes we made