GPT-5.2-Codex vs Qwen3 VL 235B A22B Instruct
OpenAI vs Alibaba — benchmarks, pricing, and capabilities side by side.
- •GPT-5.2-Codex has the higher intelligence index (89.9 vs 70.9)
- •Qwen3 VL 235B A22B Instruct is cheaper ($0.20 vs $1.75 per 1M input)
- •GPT-5.2-Codex is faster
- •GPT-5.2-Codex has a larger context window (400K)
| GPT-5.2-Codex | Qwen3 VL 235B A22B Instruct | |
|---|---|---|
| Intelligence index | 89.9 | 70.9 |
| Developer | OpenAI | Alibaba |
| Type | Multimodal | Multimodal |
| Access | API only | Open weights |
| Context window | 400,000 tokens | 262,144 tokens |
| Input price | $1.75 / 1M | $0.20 / 1M |
| Output price | $14.00 / 1M | $0.88 / 1M |
| Speed | 106 tok/s | 51 tok/s |
| Released | January 14, 2026 | September 23, 2025 |
| Parameters | — | — |
| Input modalities | Text, Image | Text, Image |
| Output modalities | Text | Text |
Shared benchmarks
GPT-5.2-Codex
Qwen3 VL 235B A22B Instruct
GPQA Diamond
89.9
71.2
Humanity’s Last Exam
33.5
6.3