OpenAI
GPT-5.5
Default frontier GPT for coding, computer use, research, and knowledge work
openai/gpt-5.5
Context
1.1M
Output
128K
Released
2026-04-23
Knowledge cutoff
2025-12-01
Types
Reasoning
Tool use
Structured output
Attachments
Open weights
Provider pricing
- Price
- $4.55 / $27.27
- Context
- 400K
- Status
- Stable
- Price
- $5.00 / $30.00
- Context
- 1M
- Status
- Stable
- Price
- $5.00 / $30.00
- Context
- 1.1M
- Status
- Stable
- Price
- $5.00 / $30.00
- Context
- 1.1M
- Status
- Stable
- Price
- $5.00 / $30.00
- Context
- 1.1M
- Status
- Stable
- Price
- $5.00 / $30.00
- Context
- 1.1M
- Status
- Stable
- Price
- $5.00 / $30.00
- Context
- 1.1M
- Status
- Stable
- Price
- $5.00 / $30.00
- Context
- 1.1M
- Status
- Stable
- Price
- $5.00 / $30.00
- Context
- 1.1M
- Status
- Stable
- Price
- $5.00 / $30.00
- Context
- 1.1M
- Status
- Stable
- Price
- $5.00 / $30.00
- Context
- 1.1M
- Status
- Stable
- Price
- $5.00 / $30.00
- Context
- 1.1M
- Status
- Stable
- Price
- $5.00 / $30.00
- Context
- 1.1M
- Status
- Stable
- Price
- $5.00 / $30.00
- Context
- 1.1M
- Status
- Stable
- Price
- $5.00 / $30.00
- Context
- 1.1M
- Status
- Stable
- Price
- $5.00 / $30.00
- Context
- 1.1M
- Status
- Stable
- Price
- $5.00 / $30.00
- Context
- 1.1M
- Status
- Stable
- Price
- $5.00 / $30.00
- Context
- 1.1M
- Status
- Stable
- Price
- $5.00 / $30.00
- Context
- 1M
- Status
- Stable
- Price
- $5.00 / $30.00
- Context
- 1.1M
- Status
- Stable
Benchmarks
SWE-Bench Pro
58.6 · resolve rate
Terminal-Bench
78.2 · success rate
SWE-Atlas Codebase QnA
45.43 · score
SWE-Atlas Refactoring
44.79 · score
SWE-Atlas Test Writing
42.59 · score
Artificial Analysis Coding Agent Index
65.3 · average pass@1
SWE-Atlas Codebase QnA
80.8 · pass@1
SWE-Bench Pro
30.9 · pass@1
Terminal-Bench
84.1 · pass@1
Artificial Analysis Coding Agent Index
60.4 · average pass@1
SWE-Atlas Codebase QnA
79.1 · pass@1
SWE-Bench Pro
26.2 · pass@1
Terminal-Bench
75.8 · pass@1
Artificial Analysis Coding Agent Index
57.8 · average pass@1
SWE-Atlas Codebase QnA
75 · pass@1
SWE-Bench Pro
24.9 · pass@1
Terminal-Bench
73.4 · pass@1
Terminal-Bench
82.7 · success rate
GPQA Diamond
93.6 · accuracy
Humanity's Last Exam
41.4 · accuracy
Humanity's Last Exam
52.2 · accuracy
OSWorld-Verified
78.7 · success rate
BrowseComp
84.4 · accuracy
MMMU Pro
81.2 · accuracy
ARC-AGI-2
85 · accuracy
FrontierMath
51.7 · accuracy
FrontierMath
35.4 · accuracy
GDPval
84.9 · wins or ties
MCP Atlas
75.3 · success rate
Toolathlon
55.6 · success rate
τ²-Bench Telecom
98 · success rate