GPT-5.5 (Non-reasoning)

GPT-5.5 (Non-reasoning)

OpenAI · Released Apr 23, 2026 Professional
Intelligence #14 / 590
82.0 Our Score
Speed #181 / 279
65.7 tokens / sec
Input #551 / 592
$5.00 per 1M tokens
Output #560 / 592
$30.00 per 1M tokens
Context
— Not reported

Analysis Summary

GPT-5.5 (Non-reasoning) is a non-thinking variant from OpenAI's GPT-5.5 family, with a coding index of 56.5 and agentic index of 59.3. Its terminal benchmark score of 0.492 and tau2 of 0.693 suggest reliable performance on structured coding and tool-use tasks, while its instruction-following and long-context retrieval scores are competitive.

At $5 input and $30 output per million tokens, it sits at a mid-range price point for the capability it delivers. It is well-suited to software development assistance, agentic pipelines, and content workflows where the full reasoning overhead of the thinking variants is unnecessary.

For teams wanting GPT-5.5 capability without the latency and cost of extended reasoning, this variant offers a practical balance for production coding and automation workloads.

Assessed June 30, 2026

Editorial notes

GPT-5.5 (Non-reasoning) from OpenAI delivers a coding index of 56.5 and agentic index of 59.3 with strong instruction following, making it a capable mid-tier option for coding and automation at $5/$30 per million tokens.

Rankings consider pricing, capabilities, benchmarks, and real-world applicability and are refreshed as new models launch. Feedback?

Performance Profile

Intelligence5Technical8.1Value4.5Content5.9
Intelligence 5/10
Technical 8.1/10
Content 5.9/10
Value 4.5/10

How GPT-5.5 (Non-reasoning) compares

GPT-5.5 (Non-reasoning) ranks #50 of 385 AI models we track for overall intelligence, #22 of 139 for coding, #64 of 293 for agentic tasks. At $5.00 per million input tokens it is cheaper than 7% of comparable models.

Performance Indices

Source: Artificial Analysis

35.4 Intelligence Index
56.5 Coding Index
59.3 Agentic Index

Benchmark Scores

Intelligence

GPQA Diamond 76.8% Graduate-level scientific reasoning
HLE 12.6% Humanity's Last Exam
SciCode 47.3% Scientific computing

Technical

TerminalBench Hard 49.2% Agentic terminal tasks
τ²-Bench 69.3% Conversational agent benchmark

Content

IFBench 46.1% Instruction following
LCR 56.3% Long-context reasoning

Benchmark data from Artificial Analysis and Hugging Face

How does GPT-5.5 (Non-reasoning) stack up?

Compare side-by-side with other professional models.

Compare Models

Model Information

ProviderOpenAI
Release Date April 23, 2026
Status Active

Pricing

Token Type Cost per 1M tokens Cost per 1K tokens
Input $5.00 $0.005000
Output $30.00 $0.030000

Leaderboard Categories

Frequently asked questions about GPT-5.5 (Non-reasoning)

How much does GPT-5.5 (Non-reasoning) cost?

GPT-5.5 (Non-reasoning) costs $5.00 per million input tokens and $30.00 per million output tokens.

Is GPT-5.5 (Non-reasoning) good for coding?

On our coding benchmark index, GPT-5.5 (Non-reasoning) ranks #22 of 139 models, placing it in the top quartile of the field for code generation and debugging.

Who created GPT-5.5 (Non-reasoning)?

GPT-5.5 (Non-reasoning) is developed by OpenAI and was released on April 23, 2026.