Updated
GPT-6.1 Sol
As of Nov 14, 2026, GPT-6.1 Sol ranks #2 of 7 on Bug Bench with a relative bug rate of 0.73 (27% fewer counted bugs than average).
- Relative bug rate
- 0.73
- Rank of 7
- #2
1.00 = average · lower is better
- Provider
- OpenAI
- Released
- Sep 29, 2026
- Price per 1M tokens
- $2 in · $10 out
- First seen
- Week of Sep 28
Nearby in the ranking
| Rank | Model | Rate |
|---|---|---|
| 1 | GPT-6 Luna | 0.31 |
| 2 | GPT-6.1 Sol | 0.73 |
| 3 | Claude Fable 5.1 | 0.88 |
| 4 | GPT-6 Astra | 0.93 |
Compared with GPT-5.6 Sol
GPT-6.1 Sol ranks #2 at 0.73; GPT-5.6 Sol is no longer ranked because it now writes under 0.5% of AI-attributed lines. GPT-6.1 Sol vs GPT-5.6 Sol
Is GPT-6.1 Sol getting worse?
Falling: 0.70 over the last 4 weeks, down from 0.91 before.
Where it does better or worse
# = rank among models
By coding agent
| Rank | Agent | Rate |
|---|---|---|
| 2 | Codex | 0.71 |
By bug type
| Rank | Bug type | Rate |
|---|---|---|
| 1 | Security and privacy | 0.60 |
| 2 | Logic bugs | 0.61 |
| 1 | Error handling | 0.64 |
| 2 | Performance | 0.98 |
| 4 | Type safety and build errors | 2.94 |
| 1 | Business rules and compliance | 0.60 |
| 2 | Architecture and design | 1.28 |
| 3 | Configuration and dependencies | 0.88 |
| 3 | Accessibility and user experience | 0.83 |
| 2 | Tests and documentation | 0.93 |
| 1 | Code maintainability | 0.69 |
| 3 | Code style | 1.46 |
By language
| Rank | Language | Rate |
|---|---|---|
| 3 | TypeScript | 0.99 |
| 1 | JavaScript | 0.52 |
| 2 | Python | 0.53 |
| 5 | Go | 1.95 |
| 1 | Rust | 0.33 |
| 1 | Swift | 0.05 |
| 1 | C# | 0.27 |
| 3 | SQL | 0.34 |
| 4 | Shell | 1.55 |
How widely GPT-6.1 Sol is used
Share of AI-written code reviewed by cubic
- Latest weekWeek of Nov 2
- 10.0%
- This week so farSince Nov 9
- 6.3%
- Peak weekWeek of Oct 19
- 12.8%
Compare GPT-6.1 Sol
Questions and answers
- Is GPT-6.1 Sol good for coding?
- GPT-6.1 Sol ranks #2 of 7 on Bug Bench, with 27% fewer counted bugs than average (0.73) from Jul 20 to Nov 8, 2026.
- Is GPT-6.1 Sol getting worse?
- Falling: 0.70 over the last 4 weeks, down from 0.91 before.
- Which coding agent has the lowest bug rate with GPT-6.1 Sol?
- Codex, at 0.71.
- How does GPT-6.1 Sol compare to GPT-5.6 Sol?
- GPT-5.6 Sol is no longer ranked because it now writes under 0.5% of AI-attributed lines.
Embed GPT-6.1 Sol's Bug Bench badge
Markdown
[](https://www.cubic.dev/bugbench/models/gpt-6-1-sol)HTML
<a href="https://www.cubic.dev/bugbench/models/gpt-6-1-sol"><img src="https://www.cubic.dev/bugbench/badge/gpt-6-1-sol" alt="GPT-6.1 Sol on Bug Bench" /></a>