Gemini 3.5 Flash-Lite: benchmarks, price and rank
Gemini 3.5 Flash-Lite from Google is 88th of 123 rated AI models on Spezia's leaderboard. It costs $0.30 in, $2.50 out per million tokens.
- 88th of 123 rated models
- Rating 917
- Rated on 10 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
- 66th of 66 on FrontierMath Tier 4
- 22nd of 24 on MRCR v2 (8 needles)
- 51st of 56 on APEX-Agents
Price: $0.30 in, $2.50 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-Agents29.3%
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.
51st of 56. Independent result from Epoch AI.
- SWE-Bench Pro54.2%
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.
37th of 50. The maker's own number, from Google, via LLM Stats.
- OSWorld-Verified74%
Ordinary tasks on a real desktop computer using normal apps: a browser, a spreadsheet, an email client, a photo editor, the file manager.
13th of 25. The maker's own number, from Google, via LLM Stats.
- FrontierMath26.0%
Hundreds of new, unpublished math problems written by mathematicians, from hard undergraduate level up to research, with answers a computer can check.
68th of 76. Independent result from Epoch AI.
- CharXiv Reasoning76.5%
Questions about real charts from scientific papers that need reasoning across the chart, not just reading off a value.
36th of 45. The maker's own number, from Google, via LLM Stats.
- MRCR v2 (8 needles)21.3%
A very long conversation full of similar requests, where the model has to find and repeat particular earlier answers, eight of them at once.
22nd of 24. The maker's own number, from Google, via LLM Stats.
- LMArena text1,435
People put a question to two anonymous models and vote for the better answer.
58th of 107. Independent result from LMArena.
- GPQA Diamond83.3%
Graduate-level questions in biology, physics and chemistry, written by PhDs so that searching the web does not give the answer away.
80th of 103. Independent result from Epoch AI.
- ARC-AGI-210.3%
Visual puzzles of colored grids, each with a new rule to work out from a few examples.
53rd of 69. Independent result from Epoch AI.
- FrontierMath Tier 40%
Research-level problems that take professional mathematicians days.
66th of 66. Independent result from Epoch AI.
1 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.