Mercury 2 is cheaper than GPT-5.3-Codex at $0.25/M vs $1.75/M input tokens.
| Spec | GPT-5.3-Codex | Mercury 2 |
|---|---|---|
| Input price | $1.75/M tokens | $0.25/M tokens |
| Output price | $14/M tokens | $0.75/M tokens |
| Context window | 400K tokens | 128K tokens |
| Released | Feb 2026 | Mar 2026 |
Not enough votes to call it. On the specs, GPT-5.3-Codex has the edge: bigger model tier, bigger context window, major provider backing.
GPT-5.3-Codex takes Web Design. Mercury 2 costs 19x less per token.
Mercury 2 is cheaper on both — 7.0× input, 19× output
Mercury 2 uses 5.1x more transitions
Mercury 2 is cheaper than GPT-5.3-Codex at $0.25/M vs $1.75/M input tokens.
| Spec | GPT-5.3-Codex | Mercury 2 |
|---|---|---|
| Input price | $1.75/M tokens | $0.25/M tokens |
| Output price | $14/M tokens | $0.75/M tokens |
| Context window | 400K tokens | 128K tokens |
| Released | Feb 2026 | Mar 2026 |
Not enough votes to call it. On the specs, GPT-5.3-Codex has the edge: bigger model tier, bigger context window, major provider backing.
GPT-5.3-Codex takes Web Design. Mercury 2 costs 19x less per token.
Mercury 2 is cheaper on both — 7.0× input, 19× output
Mercury 2 uses 5.1x more transitions