GPT-5.5
GPT-5.5 is an API-only (closed weights) language model from OpenAI, released April 23, 2026. The context window is 1.1M tokens. On the Epoch Capabilities Index it scores 160.1, rank 20 of 256 rated models. API price: $5 input / $30 output per million tokens.
Check it against your own PC →
Lab
OpenAI
Released
April 23, 2026
Weights
API-only (closed weights)
Parameters
–
Context
1.1M
Price / M tokens
$5 / $30 in / out
Epoch ECI
160.1 #20
Strengths by category
ModelTape's 0–100 fit on Epoch AI benchmark results.
| Category | Score | Based on |
|---|---|---|
| Reasoning | 36.3 | Chess Puzzles, SimpleBench, Mystery Game Puzzles |
| Math | 66.6 | ProofBench, OTIS Mock AIME 2024-2025, FrontierMath-Tiers-1-3-v2-Private, FrontierMath-Tier-4-v2-Private |
| Coding | 60.3 | GSO-Bench, MirrorCode, Terminal Bench, SWE-bench Verified, DeepSWE |
| Agent | 24.6 | Remote Labor Index, OSWorld 2.0 |
| Knowledge | 80.9 | GPQA diamond, SimpleQA Verified |
Benchmark results
| Benchmark (Epoch AI) | Score |
|---|---|
| OTIS Mock AIME | 100.0% |
| GPQA Diamond | 95.8% |
| FrontierMath (Tiers 1–3) | 89.8% |
| Terminal-Bench | 84.7% |
| SWE-bench Verified | 80.6% |
| FrontierMath-Tier-4-v2-Private | 72.5% |
| DeepSWE | 67.0% |
| SimpleQA Verified | 63.0% |
| SimpleBench | 62.8% |
| Chess Puzzles | 61.1% |
| Mystery Game Puzzles | 51.5% |
| ProofBench | 50.0% |
| GSO-Bench | 40.2% |
| EBR-bench | 34.3% |
| MirrorCode | 23.3% |
| Furniture Assembly | 20.2% |
| OSWorld 2.0 | 13.0% |
| Remote Labor Index | 6.3% |
Links
Announcement · Epoch AI benchmarks
Models with a similar score
GPT-6 Sol · GPT-5.6 Sol · GPT-5.5 Pro · Muse Wonder · Kimi K3.1 Pro · GPT-5.6 Terra · Claude Haiku 5.5 · GPT-5.4 Pro
More from OpenAI
GPT-6.1 Terra · GPT-6.1 Sol · GPT-6 Sol · GPT-6 Luna · GPT-6 Astra · GPT-5.6 Cyber · Daybreak Red · Daybreak Blue