OPENAI · MODEL TELEMETRY CARD
GPT-5.5
Canonical UID: openai/gpt-5.5-20260423 · Released: 2026-04-23
| Domain | Benchmark | Effort / Mode | ⚡ Value/Cost # | 🏆 Performance # | Score | Task $ | Unit Cost $ | Unit |
|---|---|---|---|---|---|---|---|---|
| Code | ALE-Bench | STD | #53 / 94 | #44 / 94 | 1127.58 | $0.073484 | $0.00006517 | rating point |
| Code | ALE-Bench | MEDIUM | #7 / 94 | #16 / 94 | 1589.38 | $0.328965 | $0.000207 | rating point |
| Code | ALE-Bench | XHIGH | #16 / 94 | #10 / 94 | 1942.97 | $1.51713 | $0.0007808 | rating point |
| Reason | ProofBench | XHIGH | #27 / 62 | #29 / 62 | 0.5 | $1.68179 | $3.364 | correct answer |
| Agent | DeepSWE | LOW | #59 / 65 | #61 / 65 | 0.27 | $1.2002 | $4.447 | solved task |
| Agent | DeepSWE | MEDIUM | #21 / 65 | #38 / 65 | 0.54 | $2.74934 | $5.093 | solved task |
| Agent | DeepSWE | HIGH | #34 / 65 | #28 / 65 | 0.644 | $5.10044 | $7.922 | solved task |
| Agent | DeepSWE | XHIGH | #42 / 65 | #22 / 65 | 0.67 | $7.22624 | $10.78 | solved task |
Compare against all models on ← #frontier Live Ranking Board · Fixed-capability cost half-life analysis on /methodology.