Compare/Gemini 3.1 Flash-Lite vs GPT-3.5 Turbo

Gemini 3.1 Flash-LitevsGPT-3.5 Turbo

Side-by-side comparison of pricing, 12 benchmarks, and generation speed.

Google

Gemini 3.1 Flash-Lite

Input
$0.25/M
Output
$1.5/M
Speed
303 tok/s
TTFT
5.07s
OpenAI

GPT-3.5 Turbo

Input
$0.5/M
Output
$1.5/M
Speed
129 tok/s
TTFT
0.80s

Winner by Category

Cheaper
Gemini 3.1 Flash-Lite
Faster (tok/s)
Gemini 3.1 Flash-Lite
Lower Latency
GPT-3.5 Turbo
Benchmarks (7-2)
Gemini 3.1 Flash-Lite

Pricing Comparison

MetricGemini 3.1 Flash-LiteGPT-3.5 Turbo
Input ($/M tokens)$0.25$0.5
Output ($/M tokens)$1.5$1.5
Cost for 1M input + 100K output tokens:
Gemini 3.1 Flash-Lite$0.40
GPT-3.5 Turbo$0.65

Speed Comparison

Output Speed (tokens/s) — higher is better
Gemini 3.1 Flash-Lite
303 tok/s
GPT-3.5 Turbo
129 tok/s
Time to First Token (seconds) — lower is better
Gemini 3.1 Flash-Lite
5.07s
GPT-3.5 Turbo
0.80s

Benchmark Comparison

Data from Artificial Analysis API — 12 benchmarks

Intelligence Index
25.03.6
Coding Index
34.710.7
Math Index
GPQA Diamond
82.2%29.7%
MMLU-Pro
46.2%
LiveCodeBench
AIME 2025
MATH-500
44.1%
Humanity's Last Exam
16.2%
SciCode
41.9%
IFBench
77.2%
TerminalBench
24.2%
Gemini 3.1 Flash-Lite7 wins
2 winsGPT-3.5 Turbo

Frequently Asked Questions

Which is cheaper, Gemini 3.1 Flash-Lite or GPT-3.5 Turbo?

Gemini 3.1 Flash-Lite is cheaper overall. Its blended price (3:1 input/output ratio) is $0.56/M tokens vs $0.75/M for GPT-3.5 Turbo.

Which model performs better on benchmarks?

Gemini 3.1 Flash-Lite wins 7 out of 12 benchmarks compared to 2 for GPT-3.5 Turbo. See the detailed benchmark chart above for per-category results.

Which is faster for real-time applications?

Gemini 3.1 Flash-Lite generates tokens faster at 303 tok/s vs 129 tok/s. However, GPT-3.5 Turbo has lower time-to-first-token (0.80s vs 5.07s).

When should I use Gemini 3.1 Flash-Lite vs GPT-3.5 Turbo?

Choose based on your priorities: Gemini 3.1 Flash-Lite for lower cost, Gemini 3.1 Flash-Lite for stronger benchmark performance, and Gemini 3.1 Flash-Lite for faster generation. For latency-sensitive apps, check the TTFT comparison above.