Public benchmark snapshot: 2026-07-19
ChatGPT vs Gemini: public benchmark comparison
How to read this page
This page summarizes public third-party benchmarks in plain language. It helps you see where the models stand in published data, but it does not replace trying them on your own tasks.
What public benchmarks show
Below are results only from public benchmark categories where every model on this page appears together. Scores from different sources cannot be added up, and they do not prove one model is best overall.
Arena · coding preference
Arena asks people to pick the better answer without knowing which model wrote it. Higher Elo means more people preferred that model in that category.
| Product | Exact model version | Rank | Score | Metric |
|---|---|---|---|---|
| ChatGPT | gpt-5.6-sol-xhigh (codex-harness) | 3 | 1618 | Elo (Elo) |
| Gemini | gemini-3.5-flash | 19 | 1504 | Elo (Elo) |
Arena · search-style preference
Arena asks people to pick the better answer without knowing which model wrote it. Higher Elo means more people preferred that model in that category.
| Product | Exact model version | Rank | Score | Metric |
|---|---|---|---|---|
| ChatGPT | gpt-5.5-search | 2 | 1239 | Elo (Elo) |
| Gemini | gemini-3.1-pro-grounding | 7 | 1211 | Elo (Elo) |
Arena · general chat preference
Arena asks people to pick the better answer without knowing which model wrote it. Higher Elo means more people preferred that model in that category.
| Product | Exact model version | Rank | Score | Metric |
|---|---|---|---|---|
| Gemini | gemini-3-pro | 8 | 1486 | Elo (Elo) |
| ChatGPT | gpt-5.6-sol-xhigh | 10 | 1486 | Elo (Elo) |
SWE-bench Verified · real software-issue fixing
SWE-bench Verified measures how often an AI coding setup can fix real GitHub issues. A higher resolved percentage means more issues were fixed in that test.
| Product | Exact model version | Rank | Score | Metric |
|---|---|---|---|---|
| Gemini | live-SWE-agent + Gemini 3 Pro Preview (2025-11-18) | 4 | 77.4% | Resolved (%) |
| ChatGPT | JoyCode + Claude 4 Sonnet + GPT-4.1 | 17 | 74.6% | Resolved (%) |
Product entry points
| Product | Provider | Official entry point | ModelAny |
|---|---|---|---|
| ChatGPT | OpenAI | Visit product site | Currently supported by ModelAny |
| Gemini | Visit product site | Currently supported by ModelAny |
What to check when you try them yourself
- Whether your real question is answered well, or still needs heavy editing.
- Whether the current plan is enough, and whether the price fits.
- Whether privacy, sign-in, and team features fit your needs.
More evidence-backed model comparisons
Official sources and verification date
The links below are official sources for product identity and plan information, last checked 2026-07-19. Pricing, model availability, quotas, and regional access can change; confirm on the provider site before purchasing or deploying.
- OpenAI ChatGPT plans and pricing - 2026-07-19
- Google Gemini plans - 2026-07-19
Frequently asked questions
Does this page publish an absolute ranking?
No. Conditional findings are published only when test conditions, raw outputs, and a review method are disclosed.
Are pricing and plan details definitive?
This page links to official sources and verification dates. Confirm the provider price shown for your region before purchasing.
Does ModelAny store prompts?
ModelAny is local-first. Prompts are sent only to the AI services you choose and are not uploaded to ModelAny servers.
Compare multiple models with one prompt
ModelAny is available on the Chrome Web Store. The Microsoft Edge Add-ons listing is still under review. Drafts, settings, and history remain in your browser.
Install extension