GPT-4.5 (Preview): Intelligence, Performance & Price Analysis
Analysis of OpenAI's GPT-4.5 (Preview) and comparison to other AI models across key metrics including quality, price, performance (tokens per second & time to first token), context window & more. Click on any model to compare API providers for that model. For more details including relating to our methodology, see our FAQs.
OpenAI has launched a newer model, GPT-4.1. We suggest considering this model instead of GPT-4.5 (Preview). See the following pages for a comparison of GPT-4.1 to other models and GPT-4.1 API provider benchmarks.
Comparison Summary
Intelligence:
GPT-4.5 (Preview) is of higher quality compared to average, with Intelligence Index across evaluations of 53.
Price:GPT-4.5 (Preview) is more expensive compared to average with a price of $93.75 per 1M Tokens (blended 3:1).
GPT-4.5 (Preview) Input token price: $75.00, Output token price: $150.00 per 1M Tokens.
Speed:GPT-4.5 (Preview) Input token price: $75.00, Output token price: $150.00 per 1M Tokens.
GPT-4.5 (Preview) is slower compared to average, with a output speed of 87.4 tokens per second.
Latency:GPT-4.5 (Preview) has a lower latency compared to average, taking 0.93s to receive the first token (TTFT).
Context Window:GPT-4.5 (Preview) has a smaller context windows than average, with a context window of 130k tokens.
Highlights
Intelligence
Artificial Analysis Intelligence Index; Higher is better
Estimate (independent evaluation forthcoming)
Speed
Output Tokens per Second; Higher is better
Price
USD per 1M Tokens; Lower is better
Parallel Queries:
Prompt Length:
GPT-4.5 (Preview) Model Details
Comparisons to GPT-4.5 (Preview)
GPT-4.1
o4-mini (high)
o3
Llama 4 Scout
Llama 4 Maverick
Gemini 2.5 Flash Preview (May '25) (Reasoning)
Gemini 2.5 Pro Preview (May' 25)
Claude 3.7 Sonnet (Extended Thinking)
Claude 4 Sonnet (Extended Thinking)
Claude 4 Sonnet
Mistral Medium 3
DeepSeek R1 0528 (May '25)
DeepSeek V3 0324 (Mar '25)
Grok 3 mini Reasoning (high)
Nova Premier
Llama 3.1 Nemotron Ultra 253B v1 (Reasoning)
Qwen3 235B A22B (Reasoning)
GPT-4o (Nov '24)
DeepSeek R1 (Jan '25)
Further details