GPT-5.3-Codex vs Qwen3 VL 8B Thinking
OpenAI vs Alibaba — benchmarks, pricing, and capabilities side by side.
- •Qwen3 VL 8B Thinking is cheaper ($0.12 vs $1.75 per 1M input)
- •GPT-5.3-Codex has a larger context window (400K)
| GPT-5.3-Codex | Qwen3 VL 8B Thinking | |
|---|---|---|
| Intelligence index | 91.5 | — |
| Developer | OpenAI | Alibaba |
| Type | Multimodal | Multimodal |
| Access | API only | Open weights |
| Context window | 400,000 tokens | 256,000 tokens |
| Input price | $1.75 / 1M | $0.12 / 1M |
| Output price | $14.00 / 1M | $1.37 / 1M |
| Speed | 73 tok/s | — |
| Released | February 24, 2026 | October 14, 2025 |
| Parameters | — | — |
| Input modalities | Text, Image | Image, Text |
| Output modalities | Text | Text |