OpenAI's GPT and Alibaba's Qwen, compared through their current models on the Respan Index. The strongest GPT model today is GPT-6 Astra (#2 on the Respan Index, 70.5); the strongest Qwen model is Qwen3.8-Max (#22, 50.3). GPT's best models lead in coding, agents, reasoning, knowledge and human preference.
| Category | GPT | Qwen |
|---|---|---|
| coding | GPT-6 Astra 61.8 | Qwen3.8-Max 38.9 |
| agents | GPT-6 Astra 77.0 | Qwen3.8-Max 68.0 |
| reasoning | GPT-6.1 Sol 78.6 | Qwen3.8-Max 40.8 |
| knowledge | GPT-6 Astra 75.8 | Qwen3.7-Max 55.2 |
| human preference | GPT-6 Astra 59.7 | Qwen3.8-Max 57.0 |
Each family's highest Respan Index category score; the higher one is bold. * Estimated: no published scores in that category. How the Respan Index works
Respan Index rank and score, blended list price per 1M tokens.
Respan Index rank and score, blended list price per 1M tokens.
The cheapest ranked GPT model is GPT-5 nano at $0.138 per 1M tokens blended; the cheapest Qwen model is Qwen3.5-Flash at $0.175.