Terminal-Bench 2.1 vs Effective Cost
X = effective cost (log, reversed). Y = Terminal-Bench 2.1 %. Color = provider.
Terminal-Bench 2.1 vs Effective Cost
X = effective cost (log, reversed). Y = Terminal-Bench 2.1 %. Color = provider.
How To Read This Chart
This chart uses derived AI IQ fields and source-backed benchmark data from the public AI IQ dataset.
Top Models
| Rank | Model | Provider | Score |
|---|---|---|---|
| 1 | gpt-oss-20b | OpenAI | $0.2004 |
| 2 | gpt-5-nano | OpenAI | $0.2955 |
| 3 | qwen3.5-2b | Alibaba | $0.3152 |
| 4 | llama-4-maverick | Meta | $0.3577 |
| 5 | qwen3.5-4b | Alibaba | $0.3612 |
| 6 | qwen3.5-9b | Alibaba | $0.6567 |
| 7 | mistral-large-3 | World (Other) | $0.6636 |
| 8 | mimo-v2-flash | Xiaomi | $0.7130 |
| 9 | gpt-oss-120b | OpenAI | $0.7857 |
| 10 | deepseek-v4-flash | DeepSeek | $0.8674 |
| 11 | deepseek-v3.2 | DeepSeek | $1.0026 |
| 12 | deepseek-v3.1 | DeepSeek | $1.2985 |