GPT-5.5 2026-04-23
GPT-5.5 is OpenAI’s frontier model designed for complex professional workloads, building on GPT-5.4 with stronger reasoning, higher reliability, and improved token efficiency on hard tasks. It features a 1M+ token context window (922K input, 128K output) with support for text and image inputs, enabling large-scale reasoning, coding, and multimodal workflows within a single system.
- Input / output price
- $5.00 / $30.00 per 1M
- Context
- 1.1M tokens
- Released
- Apr 23, 2026
Paste into Claude Code, Cursor, or Codex
Help me set up the Respan gateway: fetch and follow https:/
Providers
One provider serves this model, priced per 1M tokens after gateway discounts.
| Provider | Input | Output |
|---|---|---|
OpenAI openai/gpt-5.5-2026-04-23 | $5.00/M | $30.00/M |
Is a provider down? ChatGPT status
Pricing
USD per 1M tokens on OpenAI, after gateway discounts.
- Input
- $5.00 per 1M
- Output
- $30.00 per 1M
- Cache read
- $0.50 per 1M
Cost by volume
Total cost from 0 to 1T tokens.
Throughput and latency
30-day averages from requests through the Respan gateway. Under each figure, every model on the gateway by its average, this one in blue.
Output speed
Higher is better66tokens/sacross providers
Response latency
Lower is better7.4sacross providers
Benchmarks
The Respan Index plus 45 published scores from 14 sources, each linked to where it was published.
55.2/ 100
Respan Index#30 of 176
- 30GPT-5.555.2
- 28GPT-5.6 Terra55.3
- 31Gemini 3.8 Flash54.5
Published scores
| Benchmark | Score | Source | Published |
|---|---|---|---|
| WeirdMLHåvard Tveit Ihle | 84.9% | Håvard Tveit Ihle | Oct 5, 2026 |
| Terminal-Bench 2.0OpenAI | 82.7% | OpenAI | Apr 23, 2026 |
| LiveBench CodingLiveBench | 82.1% | LiveBench | Oct 5, 2026 |
| DeepSWE v1.1Datacurve | 67% | Datacurve | Oct 5, 2026 |
| SWE-Bench ProOpenAI | 58.6% | OpenAI | Apr 23, 2026 |
| GSO Opt@1GSO | 37.3% | GSO | Apr 27, 2026 |
| MirrorCodeEpoch AI | 23.3% | Epoch AI | Nov 11, 2026 |
| Benchmark | Score | Source | Published |
|---|---|---|---|
| Tau2-bench TelecomOpenAI | 98% | OpenAI | Apr 23, 2026 |
| GDPval (wins or ties)OpenAI | 84.9% | OpenAI | Apr 23, 2026 |
| BrowseCompOpenAI | 84.4% | OpenAI | Apr 23, 2026 |
| OSWorld-VerifiedOpenAI | 78.7% | OpenAI | Apr 23, 2026 |
| MCP AtlasOpenAI | 75.3% | OpenAI | Apr 23, 2026 |
| Toolathlon-VerifiedHKUST | 73.5% | HKUST | Jun 30, 2026 |
| ToolathlonOpenAI | 55.6% | OpenAI | Apr 23, 2026 |
| APEX-AgentsMercor | 55.1% | Mercor | Oct 5, 2026 |
| LiveBench Agentic CodingLiveBench | 54% | LiveBench | Oct 5, 2026 |
| tau2-bench Banking KnowledgeSierra | 44.6% | Sierra | May 5, 2026 |
| Vending-Bench 2Andon Labs | 7523.84 · #21 of 68 | Andon Labs | Oct 5, 2026 |
| Benchmark | Score | Source | Published |
|---|---|---|---|
| ARC-AGI-1 (Verified)OpenAI | 95% | OpenAI | Apr 23, 2026 |
| GPQA DiamondEpoch AI | 90.7% | Epoch AI | Oct 5, 2026 |
| LiveBench ReasoningLiveBench | 89.7% | LiveBench | Oct 5, 2026 |
| ARC-AGI-2 (Verified)OpenAI | 85% | OpenAI | Apr 23, 2026 |
| LiveBench Data AnalysisLiveBench | 81.6% | LiveBench | Oct 5, 2026 |
| SimpleBenchSimpleBench | 69% | SimpleBench | Apr 24, 2026 |
| Chess PuzzlesEpoch AI | 63% | Epoch AI | Oct 28, 2026 |
| Mystery Game PuzzlesEpoch AI | 56% | Epoch AI | Oct 5, 2026 |
| EBR-benchEpoch AI | 34.3% | Epoch AI | Oct 5, 2026 |
| Benchmark | Score | Source | Published |
|---|---|---|---|
| AIME 2026MathArena | 100% | MathArena | Sep 17, 2026 |
| LiveBench MathematicsLiveBench | 95.9% | LiveBench | Oct 5, 2026 |
| FrontierMath Tiers 1-3 v2Epoch AI | 85.3% | Epoch AI | Oct 5, 2026 |
| OTIS Mock AIME 2024-2025Epoch AI | 84.4% | Epoch AI | Oct 5, 2026 |
| FrontierMath Tier 4 v2Epoch AI | 72.5% | Epoch AI | Oct 5, 2026 |
| FrontierMath Tier 1-3OpenAI | 51.7% | OpenAI | Apr 23, 2026 |
| FrontierMath Tier 4OpenAI | 35.4% | OpenAI | Apr 23, 2026 |
| Benchmark | Score | Source | Published |
|---|---|---|---|
| LiveBench LanguageLiveBench | 87.4% | LiveBench | Oct 5, 2026 |
| LiveBench Instruction FollowingLiveBench | 70.7% | LiveBench | Oct 5, 2026 |
| SimpleQA VerifiedEpoch AI | 63% | Epoch AI | Oct 5, 2026 |
| Benchmark | Score | Source | Published |
|---|---|---|---|
| MMMU-Pro (with tools)OpenAI | 83.2% | OpenAI | Apr 23, 2026 |
| MMMU-Pro (no tools)OpenAI | 81.2% | OpenAI | Apr 23, 2026 |
| Furniture AssemblyEpoch AI | 44.2% | Epoch AI | Oct 5, 2026 |
| Benchmark | Score | Source | Published |
|---|---|---|---|
| LMArena WebDevLMArena | 1512 Elo · #61 of 124 | LMArena | Nov 20, 2026 |
| LMArena EloLMArena | 1481 Elo · #29 of 191 | LMArena | Nov 19, 2026 |
| LMArena VisionLMArena | 1285 Elo · #18 of 91 | LMArena | Nov 19, 2026 |
| LMArena AgentLMArena | 0.0471 · #20 of 68 | LMArena | Oct 12, 2026 |
How GPT compares: Claude vs GPT, Gemini vs GPT, DeepSeek vs GPT
Compare GPT-5.5 2026-04-23: GPT-5.5 vs GPT-5.4, GPT-5.5 vs GPT-6.2 Astra, GPT-5.5 vs GPT-6 Astra, GPT-5.5 vs GPT-6.1 Sol, GPT-5.5 vs GPT-6.1 Terra, GPT-5.5 vs GPT-6.2 Sol
Apps
Use GPT-5.5 2026-04-23 in the tools you already run. Each guide points the app at the Respan gateway.
Setup guides for more apps are in the integration docs.
GPT-5.5 2026-04-23 FAQ
Short answers, built from the data on this page.
GPT-5.5 2026-04-23 is a model from OpenAI, released on Apr 23, 2026. GPT-5.5 is OpenAI’s frontier model designed for complex professional workloads, building on GPT-5.4 with stronger reasoning, higher reliability, and improved token efficiency on hard tasks. It features a 1M+ token context window (922K input, 128K output) with support for text and image inputs, enabling large-scale reasoning, coding, and multimodal workflows within a single system.
On the Respan gateway, GPT-5.5 2026-04-23 costs $5.00 per 1M input tokens and $30.00 per 1M output tokens on OpenAI, after gateway discounts.
GPT-5.5 2026-04-23 is served by OpenAI.
GPT-5.5 2026-04-23 accepts up to 1.1M tokens of context.
Point any OpenAI-compatible SDK at https://gateway.respan.ai/v1 with your Respan API key and set the model to "openai/gpt-5.5-2026-04-23". To set it up with a coding agent, copy the agent setup prompt at the top of this page.
GPT-5.5 2026-04-23 scores 55.2 out of 100 on the Respan Index v1.1, ranking #30 of 176 models. By category: Coding 30.3, Agents 64.2, Reasoning 68.4, Knowledge & language 63.7, and Human preference 54.1. The Respan Index combines independent and lab-reported benchmark results into one score, using benchmark data collected by Respan.
GPT-5.5 2026-04-23 has 45 published benchmark scores, including WeirdML 84.9%, Tau2-bench Telecom 98%, and ARC-AGI-1 (Verified) 95%. Each score links to where it was published; the data is collected by Respan.
Certified where it counts.
Built to meet the security and privacy standards that enterprise and healthcare teams require.
ISO 27001
The internationally recognized standard for information security management.
SOC 2
Secure, compliant management of your data across all of our systems.
GDPR
Operated under GDPR, the world's strictest standard for data privacy.
HIPAA
HIPAA compliant, with a BAA available for healthcare teams.