OpenAI’s new AI model charges up to $1,000 per query

OpenAI’s latest AI model o3 demonstrates significant performance improvements but comes with unprecedented computational costs, reaching over $1,000 per query for complex tasks.

Key developments: OpenAI’s new o3 model employs a “test-time compute” technique that allows it to spend more time processing and exploring multiple possibilities before providing answers.

• The model achieved an impressive 87.5 percent score on the ARC-AGI benchmark, nearly triple the performance of its predecessor o1‘s 32 percent
• O3 utilizes a unique “high-compute mode” that enables more thorough analysis of complex problems
• A “mini” version of o3 is scheduled for release in January

Cost implications: The computational expenses associated with running o3 represent a significant challenge for widespread adoption and commercial viability.

• High-compute mode operations cost over $1,000 per task, approximately 170 times more than the low-compute version
• Even the low-compute version of o3 costs around $20 per task, substantially more than previous models
• These costs raise questions about sustainability given ChatGPT Plus’s $25 monthly subscription model

Performance analysis: The model’s improved capabilities demonstrate continued progress in AI development while raising questions about scalability.

• O3’s performance appears to challenge concerns about AI hitting a scaling wall
• The improvements stem from changes in reasoning methodology rather than just increased processing power
• The model approaches human-level performance on certain benchmarks, though at significantly higher costs

Expert perspective: François Chollet, creator of the ARC-AGI benchmark, provides insights into the model’s practical implications.

• Chollet notes that human workers can complete similar tasks for approximately $5 plus minimal energy costs
• Despite current cost barriers, he predicts dramatic improvements in cost-performance ratios in the coming months and years
• The model’s current performance-to-cost ratio makes it economically impractical for widespread deployment

Future implications: The substantial costs associated with o3’s operation highlight a crucial tension in AI development between performance improvements and economic feasibility.

• While o3’s capabilities represent a significant technological advancement, its high operational costs may limit practical applications
• The development suggests that future AI improvements might require balancing performance gains against computational expenses
• The success of future consumer-facing products will likely depend on finding more cost-effective ways to implement these advances