GPT-4.5
Multimodal
Zero-eval
#1CharXiv-D
#1SimpleQA
#1SWE-Lancer
+10 more
by OpenAI
About
GPT-4.5 is a multimodal language model developed by OpenAI. It achieves strong performance with an average score of 64.1% across 26 benchmarks. It excels particularly in GSM8k (97.0%), MMLU (90.8%), CharXiv-D (90.0%). The model shows particular specialization in code tasks with an average performance of 88.1%. It supports a 132K token context window for handling large documents. The model is available through 1 API provider. As a multimodal model, it can process and understand text, images, and other input formats seamlessly. Released in 2025, it represents OpenAI's latest advancement in AI technology.
Pricing Range
Input (per 1M)$75.00 -$75.00
Output (per 1M)$150.00 -$150.00
Providers1
Timeline
AnnouncedFeb 27, 2025
ReleasedFeb 27, 2025
Specifications
Capabilities
Multimodal
License & Family
License
Proprietary
Benchmark Performance Overview
Performance metrics and category breakdown
Overall Performance
26 benchmarks
Average Score
64.1%
Best Score
97.0%
High Performers (80%+)
6Performance Metrics
Max Context Window
132.1KAvg Throughput
50.0 tok/sAvg Latency
20msTop Categories
code
88.1%
math
84.7%
vision
75.2%
general
59.3%
agents
59.2%
Benchmark Performance
Top benchmark scores with normalized values (0-100%)
Ranking Across Benchmarks
Position relative to other models on each benchmark
GSM8k
Rank #3 of 46
#1o1
97.1%
#2Kimi K2 Instruct
97.3%
#3GPT-4.5
97.0%
#4Llama 3.1 405B Instruct
96.8%
#5Claude 3.5 Sonnet
96.4%
#6Claude 3.5 Sonnet
96.4%
MMLU
Rank #3 of 78
#1o1
91.8%
#2GPT-5
92.5%
#3GPT-4.5
90.8%
#4o1-preview
90.8%
#5DeepSeek-R1
90.8%
#6Claude 3.5 Sonnet
90.4%
CharXiv-D
Rank #1 of 5
#1GPT-4.5
90.0%
#2GPT-4.1 mini
88.4%
#3GPT-4.1
87.9%
#4GPT-4o
85.3%
IFEval
Rank #13 of 37
#10Llama 3.1 405B Instruct
88.6%
#11Qwen3-235B-A22B-Instruct-2507
88.7%
#12Gemma 3 12B
88.9%
#13GPT-4.5
88.2%
#14Llama 3.1 70B Instruct
87.5%
#15GPT-4.1
87.4%
#16Kimi-k1.5
87.2%
HumanEval
Rank #23 of 62
#20Claude 3.5 Haiku
88.1%
#21o1
88.1%
#22Qwen2.5 32B Instruct
88.4%
#23GPT-4.5
88.0%
#24Gemma 3 27B
87.8%
#25GPT-4o mini
87.2%
#26GPT-4 Turbo
87.1%
All Benchmark Results for GPT-4.5
Complete list of benchmark scores with detailed information
GSM8k GSM8k benchmark | math | text | 0.97 | 97.0% | Self-reported |
MMLU MMLU benchmark | general | text | 0.91 | 90.8% | Self-reported |
CharXiv-D CharXiv-D benchmark | general | text | 0.90 | 90.0% | Self-reported |
IFEval IFEval benchmark | code | text | 0.88 | 88.2% | Self-reported |
HumanEval HumanEval benchmark | code | text | 0.88 | 88.0% | Self-reported |
MMMLU MMMLU benchmark | general | text | 0.85 | 85.1% | Self-reported |
MMMU MMMU benchmark | vision | multimodal | 0.75 | 75.2% | Self-reported |
Graphwalks parents <128k Graphwalks parents <128k benchmark | general | text | 0.73 | 72.6% | Self-reported |
MathVista MathVista benchmark | math | text | 0.72 | 72.3% | Self-reported |
COLLIE COLLIE benchmark | general | text | 0.72 | 72.3% | Self-reported |
Showing 1 to 10 of 26 benchmarks