Gemini 3.6 Flash: benchmarks, price and rank
Gemini 3.6 Flash from Google is 35th of 123 rated AI models on Spezia's leaderboard. It costs $1.50 in, $7.50 out per million tokens.
- 35th of 123 rated models
- Rating 1,440
- Rated on 12 benchmarks
- Closed
- Released Jul 21, 2026
- 1 million tokens of context
- Knows the world to March 2026
Where it is strongest
Not in the top tenth on any benchmark in use.
Where it falls behind
- 48th of 66 on FrontierMath Tier 4
- 38th of 56 on APEX-Agents
- 44th of 76 on FrontierMath
Price: $1.50 in, $7.50 out per million tokens.
Its headline results
The benchmarks that count most in the rating, with its place among the models measured the same way, and where each number came from.
- APEX-Agents46.9%
Long tasks written by investment banking analysts, management consultants and corporate lawyers from their own work, done across files and apps in a realistic work setup.
38th of 56. Independent result from Epoch AI.
- SWE-Bench Pro58.7%
Long changes to real codebases in several languages, often across many files, including code from private and copyleft projects that models are unlikely to have seen.
21st of 50. The maker's own number, from Google, via LLM Stats.
- OSWorld-Verified83%
Ordinary tasks on a real desktop computer using normal apps: a browser, a spreadsheet, an email client, a photo editor, the file manager.
5th of 25. The maker's own number, from Google, via LLM Stats.
- DeepSWE 1.149%
Software engineering tasks run in the same simple agent setup for every model.
Independent result from LLM Stats.
- SimpleQA Verified66.2%
Short factual questions with one checked answer, asked without web search.
14th of 78. Independent result from Epoch AI.
- FrontierMath59.0%
Hundreds of new, unpublished math problems written by mathematicians, from hard undergraduate level up to research, with answers a computer can check.
44th of 76. Independent result from Epoch AI.
- CharXiv Reasoning89.4%
Questions about real charts from scientific papers that need reasoning across the chart, not just reading off a value.
7th of 45. The maker's own number, from Google, via LLM Stats.
- MRCR v2 (8 needles)54%
A very long conversation full of similar requests, where the model has to find and repeat particular earlier answers, eight of them at once.
10th of 24. The maker's own number, from Google, via LLM Stats.
- LMArena text1,479
People put a question to two anonymous models and vote for the better answer.
18th of 107. Independent result from LMArena.
- LMArena WebDev1,538
People ask two anonymous models to build the same web page or app and vote for the better one.
43rd of 98. Independent result from LMArena.
4 more results in its full profile.
Every result with its source, what it costs per task, what jumped with its release, side by side with any other model, and answers to your questions about when to use it.
Models ranked near it
How the rating works, and every benchmark it uses: the model leaderboard.