Claude Opus 5 vs GPT-6 Luna
Claude Opus 5 leads overall, C to D. GPT-6 Luna is about 10x cheaper per task. By suite, Claude Opus 5 takes real-world issues (C- to D+) and spec planning (C to C-). Take GPT-6 Luna anyway when you want the side that is cheaper per task and faster on real-world issues.Measured
Each agent graded on the same private suites, running headless through its own tools. Advantages are oriented so a win is a win, whichever way the metric runs.
Verdict: GPT-6 Luna vs Claude Opus 5
Claude Opus 5 leads overall, C to D. By suite: Claude Opus 5 takes real-world issues, C- to D+ (57% vs 50% capability); Claude Opus 5 takes spec planning, C to C- (66% vs 61% capability); Claude Opus 5 takes vibe coding, C+ to F (72% vs 33% capability). Claude Opus 5 used 2% of Claude Max 5x's weekly limit per pass, $0.024 per task at the listed price. GPT-6 Luna used 1% of ChatGPT Plus's weekly limit per pass, $0.002 per task at the listed price; at the ~$100 tier the board defaults to that is an estimated ~0.2% of ChatGPT Pro 5x's weekly limit (est, about $0.002 per task). Per task, GPT-6 Luna is about 10x cheaper ($0.002 vs $0.024). GPT-6 Luna is faster on real-world issues: a median task takes 1m 0s against 1m 56s. Both finished every task on it. Reasons to pick GPT-6 Luna anyway: it is cheaper per task ($0.002 vs $0.024) and it is faster on real-world issues (1m 0s vs 1m 56s median).
Changed since W32: Claude Opus 5: real-world issues usage 3% to 1% of the weekly limit.
| Dimension | Claude Opus 5 | GPT-6 Luna | Advantage |
|---|---|---|---|
| Capability | 57% | 50% | +7% (Claude Opus 5 leads) |
| Median task time | 1m 56s | 1m 0s | 56s faster (GPT-6 Luna leads) |
| Approx. task cost | $0.015 / task | $0.003 / task | $0.012 / task lower (GPT-6 Luna leads) |
Claude Opus 5: 21.0% of Claude Max 5x (21.0% / 5h, 1.0% / 1w) per run. Its weekly limit holds about 14.6 of the week's 33.6 5h windows. GPT-6 Luna: 1.0% of ChatGPT Plus (1.0% / 5h, 1.0% / 1w) per run. Its weekly limit holds about 6 of the week's 33.6 5h windows. Each figure is a share of that agent's own plan, so there is no delta to draw.
| Dimension | Claude Opus 5 | GPT-6 Luna | Advantage |
|---|---|---|---|
| Capability | 66% | 61% | +5% (Claude Opus 5 leads) |
| Median task time | 2m 9s | 1m 8s | 1m 1s faster (GPT-6 Luna leads) |
| Approx. task cost | $0.057 / task | <$0.011 / task | $0.046 / task lower (GPT-6 Luna leads) |
Claude Opus 5: 6.0% of Claude Max 5x (6.0% / 5h, 1.0% / 1w) per run. Its weekly limit holds about 14.6 of the week's 33.6 5h windows. GPT-6 Luna: under 1% of ChatGPT Plus per run. Its weekly limit holds about 6 of the week's 33.6 5h windows. Each figure is a share of that agent's own plan, so there is no delta to draw.
| Dimension | Claude Opus 5 | GPT-6 Luna | Advantage |
|---|---|---|---|
| Capability | 72% | 33% | +39% (Claude Opus 5 leads) |
| Median task time | 2m 51s | 1m 37s | 1m 15s faster (GPT-6 Luna leads) |
| Approx. task cost | $0.011 / task | not measured | - |
Claude Opus 5: 1.6% of Claude Max 5x (1.6% / 5h) per good solve. Its weekly limit holds about 14.6 of the week's 33.6 5h windows. GPT-6 Luna: under 1% of ChatGPT Plus (0.4% / 5h) per good solve. Its weekly limit holds about 6 of the week's 33.6 5h windows. Each figure is a share of that agent's own plan, so there is no delta to draw.
Grades and deltas are comparable only within a suite version. See methodology.
Frequently asked questions
Which is better for coding, Claude Opus 5 or GPT-6 Luna?
Both clear tier 2 of 5, but Claude Opus 5 scores higher on capability: 57% versus 50% on real-world@v3, our private suite of real-world coding problems. Latest measurement 2026-09-21.
Which is cheaper per task, Claude Opus 5 or GPT-6 Luna?
Claude Opus 5 used 2% of Claude Max 5x's weekly limit per pass, $0.024 per task at the listed price. GPT-6 Luna used 1% of ChatGPT Plus's weekly limit per pass, $0.002 per task at the listed price; at the ~$100 tier the board defaults to that is an estimated ~0.2% of ChatGPT Pro 5x's weekly limit (est, about $0.002 per task). Per task, GPT-6 Luna is about 10x cheaper ($0.002 vs $0.024).
What do Claude Opus 5 and GPT-6 Luna cost a month?
Claude Opus 5 runs on Claude Max 5x at $100/month (public price as of 2026-07-27). GPT-6 Luna runs on ChatGPT Plus at $20/month (public price as of 2026-07-20).