01Overview02Models03Rankings04Compare05Quiz06Guide

Model

GPT-5.5 NO.11

Previous-generation OpenAI flagship — frontier-reliable, now succeeded by GPT-5.6 Sol.

CompareBack to the index

OpenAI · GPT · Frontier · Closed weights

Specification

Fig.01 — GPT — OpenAI

USA

Context window
1.1Mtokens
Max output
128Ktokens
Input price
$5/ 1M
Output price
$30/ 1M
Cached input
$0.50/ 1M
Throughput
—
Class
Frontier
Modalities
Text, Image, Pdf
Weights
Closed
Knowledge cutoff
Dec 2025
Released
Apr 2026

Overview

GPT-5.5 was OpenAI's flagship before the July-2026 GPT-5.6 family and remains a genuinely frontier-reliable model for hard, long-horizon agentic coding — now superseded by GPT-5.6 Sol at the same price. It combines strong reasoning, elite tool use, and durable multi-step execution with a roughly 1M-token context, and stays a solid choice for teams standardized on it. New deployments should prefer GPT-5.6 Sol.

Strengths

  • Frontier long-horizon agentic reliability
  • Elite real-world coding and tool use
  • Strong reasoning across domains
  • Roughly 1M-token context window
  • Deep prompt-cache discount lowers repeat-prompt cost

Trade-offs

  • Expensive output tokens at $30/M
  • Slower than mini/nano tiers
  • >272K-token prompts incur premium long-context pricing
  • No native audio/video input

Fit

Best for

  • Autonomous coding agents and tool orchestration
  • Hard professional and engineering work
  • Complex multi-step reasoning
  • High-stakes production pipelines
  • Long-document analysis across full context

Not ideal for

  • Ultra-cheap high-volume classification
  • Latency-critical realtime use

Capabilities

Ten axes, normalized 0–100 and scored the same way across the whole catalog — so a 78 here means what a 78 means anywhere else on the bench.

Fig.02 — Capability radar

0 — 100

Reasoning
96
Coding
95
Math
95
Writing
92
Knowledge
94
Speed
55
Agentic
95
Vision
89
Multilingual
91
Long Context
92

Benchmarks

Public results, with independent third-party runs marked. Bars normalize percentages against 100 and Elo ratings against a 1500 ceiling.

Independent index

Artificial Analysis Intelligence Index

Composite of ~9–10 independent evals · Artificial Analysis

55/ 100

BenchmarkResult
Artificial Analysis Intelligence IndexIndependentGeneral · Artificial Analysis
55%
SWE-bench VerifiedIndependentCoding · SWE-bench
80.6%
SWE-bench ProIndependentCoding · SWE-bench Pro
58.6%
GPQA DiamondIndependentReasoning · Independent aggregators
94%
AIME 2025Math · Vendor-reported
95.2%
LiveBenchIndependentGeneral · LiveBench
80.7%
Terminal-Bench HardIndependentAgentic · Artificial Analysis
60.6%
DeepSWEIndependentCoding · Datacurve
70%
LMArena EloIndependentGeneral · LMArena
1506Elo

Alternatives

Models in roughly the same class — the ones worth weighing against this record.

Compare side by side
01GPT-5.6 SolOpenAIOpenAI's GPT-5.6 flagship — its best coding-and-agentic model yet, for the hardest work.02Claude Opus 4.8AnthropicState-of-the-art autonomous coding reliability at half the price of Fable 5.03GPT-5.5 ProOpenAIPrevious-generation maximum-accuracy OpenAI tier, succeeded by Sol's Ultra mode.

Index

PreviousGPT-5.6 Sol UltraNextGPT-5.5 Pro
All models/EDUCATION