OpenAI · released Apr 23, 2026
GPT-5.5
GPT-5.5 still codes and reads documents well, and newer OpenAI models now cost less.
6 months oldThinks before answering
- Input price
- $5 / M
- Output price
- $30 / M
- Context
- 1.1M tokens
- Longest reply
- 128K
- Reads
- text, image, pdf
- Knows up to
- 2025-12-01
Where it shines
- It scores 80.6 on SWE-bench and 84.7 on Terminal-Bench.
- It ranks second in the document arena at 1486.
Where it struggles
- It costs $5 in, $30 out, more than GPT-6.1 Sol.
Where you can use it
- OpenAI APIAPI$5 in / $30 out per million tokens.
- OpenRouterAPIListed as
openai/gpt-5.5
How it scores on each job
Scores run 0 to 100 and blend every measure we have for that job. The number after the dot is its rank among the 33 models in the lineup.
The evidence
| Measure | Score | Rank in lineup | Source |
|---|---|---|---|
| Epoch Capabilities Index | 160 | #9 | Epoch AI |
| LMArena text | 1472 | #14 | LMArena |
| LMArena hard prompts | 1489 | #14 | LMArena |
| LMArena instruction following | 1478 | #14 | LMArena |
| LMArena multi-turn | 1476 | #12 | LMArena |
| LMArena creative writing | 1455 | #14 | LMArena |
| LMArena writing & language | 1466 | #14 | LMArena |
| LMArena coding | 1494 | #16 | LMArena |
| LMArena math | 1487 | #15 | LMArena |
| LMArena business & finance | 1478 | #9 | LMArena |
| LMArena legal & government | 1491 | #10 | LMArena |
| LMArena medicine | 1465 | #17 | LMArena |
| Vision Arena | 1297 | #8 | LMArena |
| Search Arena | 1247 | #2 | LMArena |
| Document Arena | 1486 | #2 | LMArena |
| SWE-bench Verified | 81% | #1 | Epoch AI |
| Terminal-Bench | 85% | #1 | Epoch AI |
| GPQA Diamond | 94% | #9 | Epoch AI |
| FrontierMath | 85% | #9 | Epoch AI |
| SimpleQA Verified | 63% | #9 | Epoch AI |
| ARC-AGI-2 | 85% | #7 | Epoch AI |
| DeepResearch Bench | 54 | #1 | Epoch AI |
Compare it with these
4 months old
OpenAI
GPT-5.6 Terra
GPT-5.6 Terra is OpenAI's older mid-tier model, and GPT-6.1 Terra now costs half as much.
Long documents
- Price
- $2 in · $12 out
- Context
- 1.1M
8 months old
OpenAI
GPT-5.4
GPT-5.4 is an older OpenAI model that the GPT-6 family has passed.
Images & screenshots
- Price
- $2.5 in · $15 out
- Context
- 1.1M
New
Meta
Muse Wonder
Muse Wonder is Meta's first top-tier model, and early results show it chats nearly as well as the leaders.
Everyday questionsProfessional work
- Price
- $4 in · $20 out
- Context
- 1M