Is Qwen3 VL 235B A22B Thinking or GPT-5-Codex cheaper?

Qwen3 VL 235B A22B Thinking is cheaper on output tokens ($4.00 vs $10.00 per 1M).

Which has the larger context window, Qwen3 VL 235B A22B Thinking or GPT-5-Codex?

GPT-5-Codex has the larger context window (400K tokens).

Qwen3 VL 235B A22B Thinking vs GPT-5-Codex

Qwen3 VL 235B A22B Thinking is cheaper on output tokens, while GPT-5-Codex offers a larger context window. Choose Qwen3 VL 235B A22B Thinking or GPT-5-Codex based on the trade-off between cost, context, and the benchmarks that matter for your use case.

Spec	Qwen3 VL 235B A22B Thinking	GPT-5-Codex
Provider	OrcaRouter	OrcaRouter
Input / 1M tokens	$0.40	$1.25
Output / 1M tokens	$4.00	$10.00
Context window	131K	400K
Parameters	—	—
Open weights	Yes	No
Released	Sep 2025	Sep 2025

Qwen3 VL 235B A22B Thinking details →GPT-5-Codex details →

FAQ

Pricing is indicative — confirm with the provider before production use. Highlighted values indicate the better figure for that row.