Compare/Llama 3.2 Instruct 90B (Vision) vs Gemini 3.1 Flash-Lite Preview

Llama 3.2 Instruct 90B (Vision)vsGemini 3.1 Flash-Lite Preview

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

Meta

Llama 3.2 Instruct 90B (Vision)

Input
$1.38/M
Output
$1.38/M
Speed
45 tok/s
TTFT
0.57s
Google

Gemini 3.1 Flash-Lite Preview

Input
$0.25/M
Output
$1.5/M
Speed
284 tok/s
TTFT
5.00s

Winner by Category

Cheaper
Gemini 3.1 Flash-Lite Preview
Faster (tok/s)
Gemini 3.1 Flash-Lite Preview
Lower Latency
Llama 3.2 Instruct 90B (Vision)
Benchmarks (3-7)
Gemini 3.1 Flash-Lite Preview

Pricing Comparison

MetricLlama 3.2 Instruct 90B (Vision)Gemini 3.1 Flash-Lite Preview
Input ($/M tokens)$1.38$0.25
Output ($/M tokens)$1.38$1.5
Cost for 1M input + 100K output tokens:
Llama 3.2 Instruct 90B (Vision)$1.52
Gemini 3.1 Flash-Lite Preview$0.40

Speed Comparison

Output Speed (tokens/s) — higher is better
Llama 3.2 Instruct 90B (Vision)
45 tok/s
Gemini 3.1 Flash-Lite Preview
284 tok/s
Time to First Token (seconds) — lower is better
Llama 3.2 Instruct 90B (Vision)
0.57s
Gemini 3.1 Flash-Lite Preview
5.00s

Benchmark Comparison

Data from Artificial Analysis API — 12 benchmarks

Intelligence Index
11.933.5
Coding Index
30.1
Math Index
GPQA Diamond
43.2%82.2%
MMLU-Pro
67.1%
LiveCodeBench
21.4%
AIME 2025
MATH-500
62.9%
Humanity's Last Exam
4.9%16.2%
SciCode
24.0%41.9%
IFBench
77.2%
TerminalBench
24.2%
Llama 3.2 Instruct 90B (Vision)3 wins
7 winsGemini 3.1 Flash-Lite Preview

Frequently Asked Questions

Which is cheaper, Llama 3.2 Instruct 90B (Vision) or Gemini 3.1 Flash-Lite Preview?

Gemini 3.1 Flash-Lite Preview is cheaper overall. Its blended price (3:1 input/output ratio) is $0.56/M tokens vs $1.38/M for Llama 3.2 Instruct 90B (Vision).

Which model performs better on benchmarks?

Gemini 3.1 Flash-Lite Preview wins 7 out of 12 benchmarks compared to 3 for Llama 3.2 Instruct 90B (Vision). See the detailed benchmark chart above for per-category results.

Which is faster for real-time applications?

Gemini 3.1 Flash-Lite Preview generates tokens faster at 284 tok/s vs 45 tok/s. Llama 3.2 Instruct 90B (Vision) also has lower time-to-first-token (0.57s vs 5.00s).

When should I use Llama 3.2 Instruct 90B (Vision) vs Gemini 3.1 Flash-Lite Preview?

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