GPT-5.5vsGemini 3.1 Pro Preview
Across 11 shared benchmarks, GPT-5.5 leads overall: GPT-5.5 wins 7, Gemini 3.1 Pro Preview wins 3, with 1 ties and an average score difference of +6.36.
GPT-5.5
OpenAI · 2026-04-23 · Reasoning model
Gemini 3.1 Pro Preview
Google Deep Mind · 2026-02-20 · Multimodal model
GPT-5.57 wins(64%)Ties1(27%)3 winsGemini 3.1 Pro Preview
Benchmark scores
Grouped by capability, sorted by largest gap within each. 11 shared benchmarks.
General Knowledge
GPT-5.5 2/4| Benchmark | GPT-5.5 | Gemini 3.1 Pro Preview | Diff |
|---|---|---|---|
| ARC-AGI-2 | 851 / 59Thinking High (No Tools) | 77.107 / 59Thinking High (No Tools) | +7.90 |
| HLE | 52.2013 / 157Thinking High (With Tools) | 51.4015 / 157Thinking High (With Tools) | +0.80 |
| GPQA Diamond | 93.606 / 178Thinking High (No Tools) | 94.303 / 178Thinking High (No Tools) | -0.70 |
| ARC-AGI-3 | 02 / 6Thinking High (No Tools) | 03 / 6Thinking High (No Tools) | — |
AI Agent - Tool Usage
GPT-5.5 2/2| Benchmark | GPT-5.5 | Gemini 3.1 Pro Preview | Diff |
|---|---|---|---|
| Terminal Bench 2.0 | 82.701 / 46Thinking High (With Tools) | 68.508 / 46Thinking High (With Tools) | +14.20 |
| TerminalBench 2.1 | 83.403 / 13Thinking High (With Tools) | 70.707 / 13Thinking High (With Tools) | +12.70 |
Math and Reasoning
GPT-5.5 2/2| Benchmark | GPT-5.5 | Gemini 3.1 Pro Preview | Diff |
|---|---|---|---|
| FrontierMath - Tier 4 | 35.407 / 80Thinking High (With Tools) | 16.7020 / 80Normal (No Tools) | +18.70 |
| FrontierMath | 51.702 / 60Thinking High (With Tools) | 36.9011 / 60Thinking High (No Tools) | +14.80 |
Agent Level Benchmark
Gemini 3.1 Pro Preview 1/1| Benchmark | GPT-5.5 | Gemini 3.1 Pro Preview | Diff |
|---|---|---|---|
| τ²-Bench - Telecom | 985 / 35Thinking High (With Tools) | 99.301 / 35Thinking High (With Tools) | -1.30 |
AI Agent - Information Search
Gemini 3.1 Pro Preview 1/1| Benchmark | GPT-5.5 | Gemini 3.1 Pro Preview | Diff |
|---|---|---|---|
| BrowseComp | 84.405 / 45Thinking High (With Tools + Internet) | 85.903 / 45Thinking High (With Tools + Internet) | -1.50 |
Coding and Software Engineer
GPT-5.5 1/1| Benchmark | GPT-5.5 | Gemini 3.1 Pro Preview | Diff |
|---|---|---|---|
| SWE-Bench Pro - Public | 58.607 / 43Thinking High (With Tools) | 54.2023 / 43Thinking High (With Tools) | +4.40 |
Specs
| Field | GPT-5.5 | Gemini 3.1 Pro Preview |
|---|---|---|
| Publisher | OpenAI | Google Deep Mind |
| Release date | 2026-04-23 | 2026-02-20 |
| Model type | Reasoning model | Multimodal model |
| Architecture | Dense | Dense |
| Parameters | Not available | Not available |
| Context length | 1000K | 1M |
| Max output | 128K | 32K |
API pricing
Prices use DataLearner records when available; missing fields are not inferred.
| Item | GPT-5.5 | Gemini 3.1 Pro Preview |
|---|---|---|
| Text input | $0.5 / 1M tokens | $2 / 1M tokens |
| Text output | $30 / 1M tokens | $12 / 1M tokens |
| Cache read | $0.5 / 1M tokens | Not public |
| Cache write | $6.25 / 1M tokens | Not public |
Summary
- GPT-5.5leads in:General Knowledge (2/4), AI Agent - Tool Usage (2/2), Math and Reasoning (2/2), Coding and Software Engineer (1/1)
- Gemini 3.1 Pro Previewleads in:Agent Level Benchmark (1/1), AI Agent - Information Search (1/1)
On average across the 11 shared benchmarks, GPT-5.5 scores 6.36 higher.
Largest single-benchmark gap: FrontierMath - Tier 4 — GPT-5.5 35.40 vs Gemini 3.1 Pro Preview 16.70 (+18.70).
Page generated from structured model, pricing and benchmark records. No real-time LLM is used to write the prose.