Model Release Radar
Price/capability frontier position, benchmark scores, and community signal
Model Release Radar
GPT-5.6 Luna
openai·closed weights·Proprietary·released Thursday, Jul 9, 2026
Frontier position
- On frontier DeepSWE pass@1 (agentic coding) - no tracked model currently beats it on both price and capability.
Cost vs. capability
Loading chart…
Benchmark scores
Measured cost per task (DeepSWE)
| Benchmark | Pass@1 | Measured cost / task | Frontier |
|---|---|---|---|
| DeepSWE (agentic coding) (max effort) | 67.2% | $0.606 | frontier |
Artificial Analysis scores (no matched cost)
Artificial Analysis scores below have no measured per-task cost, so no frontier claim is made for them here.
| Benchmark | Score |
|---|---|
| AA intelligence index | 50.1 |
| AA coding index | 68.6 |
| GPQA Diamond (science) | 89.5% |
| Humanity's Last Exam | 37.0% |
| LCR | 73.3% |
| SciCode (coding) | 50.0% |
| Tau2 (banking) | 28.7% |
| Terminal-Bench 2.1 (agentic) | 77.9% |
Variants
This page covers 6 reasoning-effort variants of the same model.
| Variant | AA coding index | AA intelligence index | Price /1M |
|---|---|---|---|
| Extra-high effort (shown above) | 68.6 | 50.1 | $0.45/1M |
| High effort | 63.3 | 47.0 | $0.45/1M |
| Low effort | 44.2 | 33.9 | $0.45/1M |
| Max effort | 71.4 | 52.3 | $0.45/1M |
| Medium effort | 50.7 | 38.9 | $0.45/1M |
| Non-reasoning | 39.3 | 26.8 | $0.45/1M |
Community signal
| Variant | Coding Elo | Overall Elo | Votes |
|---|---|---|---|
| Extra-high effort | 1,462 | 1,429 | 22,852 |