| Claude Opus 5 | GPT-6 Luna | |
|---|---|---|
| Input, per 1M tokens | $5.00 | $0.10 |
| Output, per 1M tokens | $25.00 | $0.50 |
| Context window | 1M | 1.05M |
| Classification or routing (600 tokens in, 10 out), per 1,000 | $3.25 | $0.065 |
| Chat reply (2,000 in, 400 out), per 1,000 | $20.00 | $0.40 |
| Retrieval-augmented answer (8,000 in, 500 out), per 1,000 | $52.50 | $1.05 |
| Confidence signal | no | no |
| Tool calling | yes | yes |
| BANKING77, correct of 400 | not measured | 320 |
Which one should carry your traffic?
If GPT-6 Luna answers your requests as well as Claude Opus 5, moving that traffic saves 98% on chat-shaped requests. Averages cannot tell you whether it does for your requests. GPT-6 Luna does not expose a confidence signal through OpenRouter, so Gimbix would test it as a full replacement rather than a first answer.
Claude Opus 5 profile · GPT-6 Luna profile · Measure it on your traffic