GPT-6.1 Sol effort guide: ★ Med recommended, ability and cost by effort
GPT-6.1 Sol (OpenAI): the recommended effort is ★ Med, ability 51.2 at $0.613 per reference coding task. One step up to High: +2.6 ability for ×1.39 the cost.
Ability, cost and time at each effort
GPT-6.1 Sol · OpenAI · ★ Med
| Effort | Ability | vs ★ | Cost / task | Cost × | Time × |
|---|---|---|---|---|---|
| Low | 42.4 | -8.8 | $0.414 | ×0.68 | ×0.80 |
| ★ Med | 51.2 | — | $0.613 | ×1.00 | ×1.00 |
| High | 53.8 | +2.6 | $0.852 | ×1.39 | ×1.23 |
| XHigh | 56.1 | +4.9 | $1.01 | ×1.65 | ×1.43 |
| Max | 55.5 | +4.3 | $1.44 | ×2.35 | ×2.25 |
- ⚡ Fast (Codex): plan-quota ×2.5, credits ×2; speed-up not stated
- ⚡ Ultrafast (Codex): plan-quota ×8, credits ×6; speed-up not stated
Cost is the API cost of one reference coding task; cost × and time × are relative to the ★ recommended effort. Open in the effort console
In OpenAI's lineup
Strongest · Best value
- GPT-6 Luna → GPT-6.1 Sol: ability +15.9, cost ×3.65
Where to use it
- API:
gpt-6.1-sol - Codex:
gpt-6.1-sol
Other models from this maker: GPT-6 Astra · GPT-6 Luna · GPT-5.6 Sol · GPT-5.6 Terra
Data published 2026-10-15 11:33 UTC · All data is also available through the public API