New

Work in progress: Agents Directory has just launched. Stay tuned, more content is on the way.

GPT-5.5

OpenAI's current flagship, with strong agentic coding and good cost efficiency.

Capabilities:

Input

Text input
Image input (vision)
File input (PDF)
Audio input
Video input

Output

Text output
Image output
Audio output

Benchmark scores:

BenchmarkRankScore

CursorBench 3.1
Rank#3 of 10
Score64.3%
Artificial Analysis Intelligence Index
Rank#3 of 19
Score60.2
FrontierCode Main
Rank#3 of 12
Score25.5%
FrontierCode Diamond
Rank#2 of 11
Score6.3%

GPT-5.5 on CursorBench 3.1

64.3%

Ambiguous, multi-file tasks from real Cursor sessions that test codebase understanding, bugfinding, planning, and code review.

GPT-5.5 (Extra High) scores 64.3%, #3 of 10 tracked models.
Claude Fable 5 leads at 72.9%. The group median is 57.7%.
Also tested: High 62.6%, Medium 59.2%, Low 48.8%.

vs. top models

Full interactive leaderboard on our CursorBench 3.1 page.

Source: cursor.com · Updated June 2026

GPT-5.5 on Artificial Analysis Intelligence Index

60.2

The most-cited composite intelligence score: a 0–100 index combining knowledge, reasoning, math, coding, and agentic evaluations (GPQA Diamond, HLE, IFBench, SciCode, Terminal-Bench Hard, τ²-Bench, and more). Higher is better.

GPT-5.5 (xhigh) scores 60.2, #3 of 19 tracked models.
Claude Fable 5 leads at 64.9. The group median is 53.2.
Also tested: high 58.9, medium 56.7, low 50.8, May 2026 41.8, Non-reasoning 40.9.

Source: artificialanalysis.ai · Updated June 2026

GPT-5.5 on FrontierCode Main

25.5%

Cognition's test of whether a model writes code maintainers would actually merge, not just code that passes tests. Main is the 100 hardest of 150 tasks. Higher is better.

GPT-5.5 scores 25.5%, #3 of 12 tracked models.
Claude Fable 5 leads at 46.3%. The group median is 16.35%.

Source: cognition.ai · Updated June 2026

Fast mode:

Output speed
Up to 1.5x faster
Fast mode price
$12.5 / $75 per M
Standard price
$5 / $30 per M
Price premium
up to 2.5x

Pricing & availability:

OpenRouter
$5 / $30 per M
Azure AI Foundry
$5 / $30 per M
OpenAI API
$5 / $30 per M

Appears in rankings:

Best models for Codex

Strengths:

agentic codingreasoningskill use

Sources:

OpenRouter

Share:

Details:

Provider
OpenAI
Context window
1.05M
Input price
$5/M
Output price
$30/M
Open weights
No
Knowledge cutoff
Dec 2025
Released
Apr 2026

GPT-5.5

OpenAI's current flagship, with strong agentic coding and good cost efficiency.

Capabilities:

GPT-5.5 on CursorBench 3.1

GPT-5.5 on Artificial Analysis Intelligence Index

GPT-5.5 on FrontierCode Main

GPT-5.5 on CursorBench 3.1

GPT-5.5 on Artificial Analysis Intelligence Index

GPT-5.5 on FrontierCode Main

GPT-5.5 on FrontierCode Diamond

GPT-5.5 on FrontierCode Diamond