Compare/DeepSeek V3.2 Exp (Reasoning) vs Gemini 2.5 Flash-Lite Preview (Sep '25) (Non-reasoning)

DeepSeek V3.2 Exp (Reasoning)vsGemini 2.5 Flash-Lite Preview (Sep '25) (Non-reasoning)

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

DeepSeek

DeepSeek V3.2 Exp (Reasoning)

Input
$0.28/M
Output
$0.42/M
Speed
32 tok/s
TTFT
1.48s
Google

Gemini 2.5 Flash-Lite Preview (Sep '25) (Non-reasoning)

Input
$0.1/M
Output
$0.4/M
Speed
344 tok/s
TTFT
0.36s

Winner by Category

Cheaper
Gemini 2.5 Flash-Lite Preview (Sep '25) (Non-reasoning)
Faster (tok/s)
Gemini 2.5 Flash-Lite Preview (Sep '25) (Non-reasoning)
Lower Latency
Gemini 2.5 Flash-Lite Preview (Sep '25) (Non-reasoning)
Benchmarks (11-0)
DeepSeek V3.2 Exp (Reasoning)

Pricing Comparison

MetricDeepSeek V3.2 Exp (Reasoning)Gemini 2.5 Flash-Lite Preview (Sep '25) (Non-reasoning)
Input ($/M tokens)$0.28$0.1
Output ($/M tokens)$0.42$0.4
Cost for 1M input + 100K output tokens:
DeepSeek V3.2 Exp (Reasoning)$0.32
Gemini 2.5 Flash-Lite Preview (Sep '25) (Non-reasoning)$0.14

Speed Comparison

Output Speed (tokens/s) — higher is better
DeepSeek V3.2 Exp (Reasoning)
32 tok/s
Gemini 2.5 Flash-Lite Preview (Sep '25) (Non-reasoning)
344 tok/s
Time to First Token (seconds) — lower is better
DeepSeek V3.2 Exp (Reasoning)
1.48s
Gemini 2.5 Flash-Lite Preview (Sep '25) (Non-reasoning)
0.36s

Benchmark Comparison

Data from Artificial Analysis API — 12 benchmarks

Intelligence Index
32.919.4
Coding Index
33.314.5
Math Index
87.746.7
GPQA Diamond
79.7%65.1%
MMLU-Pro
85.0%79.6%
LiveCodeBench
78.9%64.1%
AIME 2025
87.7%46.7%
MATH-500
Humanity's Last Exam
13.8%4.6%
SciCode
37.7%28.5%
IFBench
54.1%41.8%
TerminalBench
31.1%7.6%
DeepSeek V3.2 Exp (Reasoning)11 wins
0 winsGemini 2.5 Flash-Lite Preview (Sep '25) (Non-reasoning)

Frequently Asked Questions

Which is cheaper, DeepSeek V3.2 Exp (Reasoning) or Gemini 2.5 Flash-Lite Preview (Sep '25) (Non-reasoning)?

Gemini 2.5 Flash-Lite Preview (Sep '25) (Non-reasoning) is cheaper overall. Its blended price (3:1 input/output ratio) is $0.17/M tokens vs $0.32/M for DeepSeek V3.2 Exp (Reasoning).

Which model performs better on benchmarks?

DeepSeek V3.2 Exp (Reasoning) wins 11 out of 12 benchmarks compared to 0 for Gemini 2.5 Flash-Lite Preview (Sep '25) (Non-reasoning). See the detailed benchmark chart above for per-category results.

Which is faster for real-time applications?

Gemini 2.5 Flash-Lite Preview (Sep '25) (Non-reasoning) generates tokens faster at 344 tok/s vs 32 tok/s. However, Gemini 2.5 Flash-Lite Preview (Sep '25) (Non-reasoning) has lower time-to-first-token (0.36s vs 1.48s).

When should I use DeepSeek V3.2 Exp (Reasoning) vs Gemini 2.5 Flash-Lite Preview (Sep '25) (Non-reasoning)?

Choose based on your priorities: Gemini 2.5 Flash-Lite Preview (Sep '25) (Non-reasoning) for lower cost, DeepSeek V3.2 Exp (Reasoning) for stronger benchmark performance, and Gemini 2.5 Flash-Lite Preview (Sep '25) (Non-reasoning) for faster generation. For latency-sensitive apps, check the TTFT comparison above.