The cost of AI reasoning over time (via) Karina Nguyen from Anthropic provides a fascinating visualization illustrating the cost of different levels of LLM over the past few years, plotting their cost-per-token against their scores on the MMLU benchmark.
Claude 3 Haiku currently occupies the lowest cost to score ratio, over on the lower right hand side of the chart.
Recent articles
- Trying out Gemini 3 Pro with audio transcription and a new pelican benchmark - 18th November 2025
- What happens if AI labs train for pelicans riding bicycles? - 13th November 2025
- Reverse engineering Codex CLI to get GPT-5-Codex-Mini to draw me a pelican - 9th November 2025