GPT-5.4 (Non-reasoning)
- Released
- Mar 5, 2026
- Tracked since
- 2026-07-23
Intelligence index18.2Ranks 232 of 664 models with this metric
Coding index—Not tested
Agentic index—Not tested
Capability profile
5 axes vs. site averageGPT-5.4 (Non-reasoning)
Site average
Some dimensions have no data yet — marked on the chart
Pricing
per 1M tokens- Input
- $2.50
- Output
- $15.00
- Cache read
- $0.250
Performance
- Output throughput
- 96 tokens/sec
- Time to first token
- 860ms
- First answer token
- 860ms
- End-to-end response
- 6.08s
History
2Intelligence index
27.7→18.2
Output
Unchanged across 62 days of records
Output throughput
96→96
Cheaper at this level
Models with a comparable intelligence index and a lower output price- DeepSeek V4 Flash 0420 (Non-reasoning)
DeepSeekIntelligence 18.9$0.200−99%
- Hy3-preview (Reasoning)
TencentIntelligence 22.7$0.210−99%
- Ling 3.0 Flash
InclusionAIIntelligence 24.9$0.220−99%
- Ling-3.0-flash-VL
InclusionAIIntelligence 24.6$0.220−99%
- Ling-3.0-flash-Fin
InclusionAIIntelligence 22.6$0.220−99%