# GPT 6 Luna max vs Muse Spark 1.3 high: real-repository coding comparison

Stet publishes a public search API over its indexed pages; the guide is at https://www.stet.sh/agents/agent-access.

Source: https://www.stet.sh/leaderboard/compare/gpt-6-luna-max-vs-muse-spark-1-3-high
Source observed: 2026-11-19T11:08:45.959Z
Last checked: 2026-11-19T11:08:45.959+00:00
Indexed: 2026-11-19T11:11:14.137+00:00

---

# GPT 6 Luna max vs Muse Spark 1.3 high

Muse Spark 1.3 high leads on speed and cost, with no clear quality difference in this comparison. GPT 6 Luna max used the `codex` harness with maximum reasoning; Muse Spark 1.3 high used `muse` with high reasoning. We compare these complete setups, not models in isolation.

## At a glance

| Dimension | Result |
|---|---|
| Quality | Even; no clear difference: GPT 6 Luna max won 3 head-to-head tasks, Muse Spark 1.3 high won 4, 15 tied, and 2 were undecided |
| Speed | Muse Spark 1.3 high was 4.5× faster: 13.2 min vs 59.7 min per task |
| Cost | Muse Spark 1.3 high was 3.7× cheaper: $0.10 vs $0.37 per task |

## What we compared

- **One real repository:** private `Stet-AI/stet`; these are not public benchmark scores.
- **24 merged changes:** each setup ran end to end from the repository state before the change.
- **Grading against the merged change:** the repository’s tests and judges comparing each patch with the merged change.
- **Quality, cost, and time:** a smaller sample than public benchmarks, but real agent work on a real codebase.

Data: November 19, 2026. See [GPT 6 Luna vs Muse Spark 1.3: all effort pairings](https://www.stet.sh/leaderboard/compare/gpt-6-luna-vs-muse-spark-1-3).

## Stet score

| Setup | Harness and reasoning | Rank | Stet score |
|---|---|---:|---:|
| GPT 6 Luna max | `codex` · max reasoning | 15 of 18 | 45.395% (95% range 39.3–51.4) |
| Muse Spark 1.3 high | `muse` · high reasoning | 11 of 18 | 47.595% (95% range 41.6–52.9) |

The 95% score ranges overlap. A Stet score is a setup’s average chance of beating another setup on a task, from 0 to 100; 50 is even with the field.

## Quality

Head-to-head results were GPT 6 Luna max 3 wins, 15 ties, 2 undecided, and Muse Spark 1.3 high 4 wins. Muse Spark 1.3 high’s win share was 51%, with a 95% range of 39–63%. Ties are within judge noise.

### Other quality measures

| Metric | GPT 6 Luna max | Muse Spark 1.3 high |
|---|---:|---:|
| Code review pass rate | 25% (24 tasks) | 24% (21 tasks) |
| Checklist score | 0.89 (24 tasks) | 0.78 (24 tasks) |
| Equivalence rate | 46% (24 tasks) | 38% (21 tasks) |

These measures are not part of the headline Stet score and have no confidence interval; a small gap may be noise. The higher recorded value is not a proven lead.

## Task by task

GPT 6 Luna max solved 2 tasks that Muse Spark 1.3 high did not; Muse Spark 1.3 high solved 2 that GPT 6 Luna max did not. Both solved 9, and neither solved 9. On the 9 tasks both solved, judges preferred Muse Spark 1.3 high’s patch on 2 and called 7 even.

### Only GPT 6 Luna max solved

- 113 changed lines in 4 files — `stet-commit-f0caded9`; judges preferred GPT 6 Luna max’s patch.
- 640 changed lines in 8 files — `stet-commit-342d0ac0`; judges preferred GPT 6 Luna max’s patch.

### Only Muse Spark 1.3 high solved

- 2 changed lines in 1 file — `stet-2450ca2d`; judges preferred Muse Spark 1.3 high’s patch.
- 1,312 changed lines in 9 files — `stet-commit-f8ae1177`; judges preferred Muse Spark 1.3 high’s patch.

### Both solved

- 4 changed lines in 1 file — `stet-43493c52`.
- 6 changed lines in 1 file — `stet-3169aefa`.
- 9 changed lines in 1 file — `stet-89dfbc27`.
- 10 changed lines in 1 file — `stet-232b50ba`.
- 23 changed lines in 2 files — `stet-5276aa3f`.
- 29 changed lines in 1 file — `stet-b28f4fe7`.
- 84 changed lines in 1 file — `stet-66762b4c`; judges preferred Muse Spark 1.3 high’s patch.
- 218 changed lines in 2 files — `stet-commit-5d766dde`; judges preferred Muse Spark 1.3 high’s patch.
- 941 changed lines in 7 files — `stet-commit-bbbbae09`.

### Neither solved

- 5 changed lines in 2 files — `stet-15439c21`.
- 23 changed lines in 1 file — `stet-6f84e978`.
- 149 changed lines in 2 files — `stet-commit-e928166f`; judges preferred GPT 6 Luna max’s patch.
- 476 changed lines in 17 files — `stet-commit-30506f8e`.
- 481 changed lines in 3 files — `stet-commit-5e075641`.
- 552 changed lines in 11 files — `stet-commit-60a0e6c2`.
- 564 changed lines in 10 files — `stet-commit-a8716b88`.
- 1,236 changed lines in 16 files — `stet-commit-d2ab0c25`.
- 1,420 changed lines in 19 files — `stet-commit-2826765e`.

A task counts as solved when its tests passed or a judge found the fix equivalent to the merged change. Stet score also weighs code quality, so it can order setups differently. Two of the 24 tasks are left out of the detailed comparison because one setup has no scored result on them.

## All 22 compared tasks

| Task | GPT 6 Luna max | Muse Spark 1.3 high | Solved by | Better patch |
|---|---|---|---|---|
| `stet-89dfbc27` · 9 lines, 1 file | Passed tests · $0.03 · 9.0 min | Passed tests · $0.01 · 2.4 min | Both | Even |
| `stet-b28f4fe7` · 29 lines, 1 file | Passed tests · $0.03 · 7.2 min | Passed tests · $0.03 · 3.4 min | Both | Even |
| `stet-232b50ba` · 10 lines, 1 file | Passed tests · $0.05 · 10.1 min | Passed tests · $0.02 · 2.1 min | Both | Even |
| `stet-3169aefa` · 6 lines, 1 file | Passed tests · $0.09 · 16.9 min | Passed tests · $0.04 · 2.5 min | Both | Even |
| `stet-5276aa3f` · 23 lines, 2 files | Passed tests · $0.04 · 9.4 min | Passed tests · $0.03 · 3.4 min | Both | Even |
| `stet-66762b4c` · 84 lines, 1 file | Equivalent to merged fix · $0.07 · 14.7 min | Passed tests · $0.02 · 4.1 min | Both | Muse Spark 1.3 high |
| `stet-commit-bbbbae09` · 941 lines, 7 files | Passed tests · $0.55 · 92.9 min | Passed tests · $0.16 · 33.7 min | Both | Even |
| `stet-commit-5d766dde` · 218 lines, 2 files | Equivalent to merged fix · $0.26 · 38.4 min | Equivalent to merged fix · $0.05 · 8.9 min | Both | Muse Spark 1.3 high |
| `stet-43493c52` · 4 lines, 1 file | Passed tests · $0.19 · 36.4 min | Passed tests · $0.06 · 15.3 min | Both | Even |
| `stet-commit-f0caded9` · 113 lines, 4 files | Equivalent to merged fix · $0.33 · 59.4 min | Failed · $0.03 · 3.3 min | GPT 6 Luna max only | GPT 6 Luna max |
| `stet-15439c21` · 5 lines, 2 files | Failed · $0.10 · 19.3 min | Failed · $0.02 · 1.6 min | Neither | Even |
| `stet-commit-f8ae1177` · 1,312 lines, 9 files | Failed · $0.68 · 109.7 min | Passed tests · $0.18 · 32.4 min | Muse Spark 1.3 high only | Muse Spark 1.3 high |
| `stet-2450ca2d` · 2 lines, 1 file | Failed · $0.06 · 15.8 min | Equivalent to merged fix · $0.02 · 3.4 min | Muse Spark 1.3 high only | Muse Spark 1.3 high |
| `stet-commit-342d0ac0` · 640 lines, 8 files | Equivalent to merged fix · $0.44 · 75.2 min | Failed · $0.19 · 20.8 min | GPT 6 Luna max only | GPT 6 Luna max |
| `stet-6f84e978` · 23 lines, 1 file | Failed · $0.09 · 20.7 min | Failed · $0.02 · 3.1 min | Neither | Even |
| `stet-commit-5e075641` · 481 lines, 3 files | Failed · $0.71 · 113.8 min | Failed · $0.07 · 12.2 min | Neither | Even |
| `stet-commit-60a0e6c2` · 552 lines, 11 files | Failed · $0.48 · 77.7 min | Failed · $0.17 · 15.6 min | Neither | Even |
| `stet-commit-2826765e` · 1,420 lines, 19 files | Failed · $0.79 · 118.1 min | Failed · $0.19 · 33.4 min | Neither | Even |
| `stet-commit-a8716b88` · 564 lines, 10 files | Failed · $0.43 · 79.0 min | Failed · $0.09 · 17.0 min | Neither | Even |
| `stet-commit-e928166f` · 149 lines, 2 files | Failed · $0.26 · 49.7 min | Failed · $0.04 · 6.2 min | Neither | GPT 6 Luna max |
| `stet-commit-30506f8e` · 476 lines, 17 files | Failed · $0.60 · 95.9 min | Failed · $0.30 · 22.9 min | Neither | Even |
| `stet-commit-d2ab0c25` · 1,236 lines, 16 files | Failed · $0.58 · 87.1 min | Failed · $0.46 · 38.1 min | Neither | Even |

## Cost and time

On the 22 compared tasks, Muse Spark 1.3 high cost less on 21; GPT 6 Luna max cost less on 1.

- **Cost per task:** GPT 6 Luna max mean $0.31, median $0.26; Muse Spark 1.3 high mean $0.10, median $0.05.
- **Agent time per task:** GPT 6 Luna max mean 52.6 min, median 44.1 min; Muse Spark 1.3 high mean 13.0 min, median 7.6 min. Muse Spark 1.3 high finished faster on all 22 tasks.

## Where each setup may fit

These groups contain fewer tasks than the overall standings and are leads to test on your repository.

### Small fixes

Ten tasks with 2–84 changed lines across 1–2 files:

| Metric | GPT 6 Luna max | Muse Spark 1.3 high |
|---|---:|---:|
| Solved | 7/10 | 8/10 |
| Cost per task | $0.07 | $0.03 |
| Agent time per task | 16.0 min | 4.1 min |

Muse Spark 1.3 high solved more at 2.8× lower cost per task and was 3.9× faster.

### Large, cross-cutting changes

Fourteen tasks with 113–1,420 changed lines across up to 19 files:

| Metric | GPT 6 Luna max | Muse Spark 1.3 high |
|---|---:|---:|
| Solved | 4/12 | 3/12 |
| Cost per task | $0.51 | $0.16 |
| Agent time per task | 83.1 min | 20.4 min |

GPT 6 Luna max solved one more task; Muse Spark 1.3 high cost 3.1× less per task and was 4.1× faster.

## Side by side

| Metric | GPT 6 Luna max | Muse Spark 1.3 high |
|---|---|---|
| Harness | `codex` | `muse` |
| Reasoning effort | max | high |
| Rank | 15 of 18 | 11 of 18 |
| Stet score | 45.395%, range 39.3–51.4 | 47.595%, range 41.6–52.9 |
| Solved | 11/24: 7 passed, 4 judged equivalent | 11/24: 9 passed, 2 judged equivalent |
| Cost per task | $0.37 | $0.10 |
| Agent time per task | 59.7 min | 13.2 min |

## Research links

- [Cost per successful task ranked the coding agent that lost every head to head 1st](https://www.stet.sh/blog/cost-per-successful-task)
- [A benchmark mined from my own repo agreed with FrontierCode](https://www.stet.sh/blog/benchmark-agreed-with-frontier-code)
- [I ran GPT 6.1 Sol against GPT 6 Sol on 24 tasks from my own repo, and found that it maintained quality while cutting cost in half](https://www.stet.sh/blog/gpt-6-1-sol-vs-gpt-6-sol)

## Questions

**Which is better for coding?** There is no clear quality difference: Muse Spark 1.3 high won 4, GPT 6 Luna max won 3, 15 tied, and 2 were undecided. Muse Spark 1.3 high led on cost and agent time: 3.7× cheaper and 4.5× faster, measured on 24 merged changes from `Stet-AI/stet`.

**How did we test them?** Both coding agents worked on the same 24 merged changes from the private `Stet-AI/stet` repository, each from the real repository state before the change. The repository’s tests and judges graded every patch against the merged change; we used no public benchmark scores.

## Test your own repository

We replay your merged changes, score quality above pass/fail, and deliver comparison reports through your agent. [Get started](https://www.stet.sh/private)

## About this data

- 24 tasks, each a merged change from the private `Stet-AI/stet` repository replayed from the state immediately before the change.
- Data from November 19, 2026.
- Both setups came from leaderboard run `bmk_d4ba310fdf50db4ccb2092e3`, containing 18 setups.

[See all 18 setups on the leaderboard](https://www.stet.sh/leaderboard) · [How Stet measures](https://www.stet.sh/methodology)

## More comparisons

- [GPT 6 Luna max vs GPT 5.6 Luna max](https://www.stet.sh/leaderboard/compare/gpt-5-6-luna-max-vs-gpt-6-luna-max)
- [GPT 6 Luna max vs Haiku 5.5 low](https://www.stet.sh/leaderboard/compare/gpt-6-luna-max-vs-haiku-5-5-low)
- [GPT 6 Luna max vs Opus 4.8 medium](https://www.stet.sh/leaderboard/compare/gpt-6-luna-max-vs-opus-4-8-medium)
- [GPT 6 Luna max vs GPT 5.6 Luna low](https://www.stet.sh/leaderboard/compare/gpt-5-6-luna-low-vs-gpt-6-luna-max)
- [GPT 6 Luna max vs GLM 5.3 Flash high](https://www.stet.sh/leaderboard/compare/glm-5-3-flash-high-vs-gpt-6-luna-max)
- [Muse Spark 1.3 high vs Haiku 5.5 max](https://www.stet.sh/leaderboard/compare/haiku-5-5-max-vs-muse-spark-1-3-high)
- [Muse Spark 1.3 high vs GLM 5.3 Flash high](https://www.stet.sh/leaderboard/compare/glm-5-3-flash-high-vs-muse-spark-1-3-high)
- [Muse Spark 1.3 high vs DeepSeek V4.1 Flash (10/17) max](https://www.stet.sh/leaderboard/compare/deepseek-v4-1-flash-10-17-max-vs-muse-spark-1-3-high)
- [Muse Spark 1.3 high vs Opus 4.8 medium](https://www.stet.sh/leaderboard/compare/muse-spark-1-3-high-vs-opus-4-8-medium)
- [Muse Spark 1.3 high vs DeepSeek V4.1 Flash max](https://www.stet.sh/leaderboard/compare/deepseek-v4-1-flash-max-vs-muse-spark-1-3-high)
- [All 153 comparisons](https://www.stet.sh/leaderboard/compare)

---

[Read this page’s summary](https://www.stet.sh/agents/api/public/v1/page-tldr?url=https%3A%2F%2Fwww.stet.sh%2Fleaderboard%2Fcompare%2Fgpt-6-luna-max-vs-muse-spark-1-3-high&trace=t_5da45423e484)

## Ask a follow-up about this page

Need a more specific answer? Ask the exact question you have. It returns a fresh, cited answer from Stet’s knowledge base. No account or API key required.

Open this URL with your own question, URL-encoded. Include your question in q. All research inputs are optional; no account or setup.

`GET https://www.stet.sh/agents/agent/faq?q={question}&trace=t_5da45423e484`

[Tool guide and examples](https://www.stet.sh/agents/agent-access) · [OpenAPI](https://www.stet.sh/agents/api/public/v1/openapi.json) · [Page TLDR form](https://www.stet.sh/agents/agent-tools/page-tldr?url=https%3A%2F%2Fwww.stet.sh%2Fleaderboard%2Fcompare%2Fgpt-6-luna-max-vs-muse-spark-1-3-high). Opening a form does not execute a tool.