Compare/Llama 3.2 Instruct 90B (Vision) vs Qwen3 8B (Non-reasoning)

Llama 3.2 Instruct 90B (Vision)vsQwen3 8B (Non-reasoning)

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

Meta

Llama 3.2 Instruct 90B (Vision)

Input
$0.72/M
Output
$0.72/M
Speed
42 tok/s
TTFT
0.37s
Alibaba

Qwen3 8B (Non-reasoning)

Input
$0.18/M
Output
$0.7/M
Speed
81 tok/s
TTFT
0.95s

Winner by Category

Cheaper
Qwen3 8B (Non-reasoning)
Faster (tok/s)
Qwen3 8B (Non-reasoning)
Lower Latency
Llama 3.2 Instruct 90B (Vision)
Benchmarks (5-7)
Qwen3 8B (Non-reasoning)

Pricing Comparison

MetricLlama 3.2 Instruct 90B (Vision)Qwen3 8B (Non-reasoning)
Input ($/M tokens)$0.72$0.18
Output ($/M tokens)$0.72$0.7
Cost for 1M input + 100K output tokens:
Llama 3.2 Instruct 90B (Vision)$0.79
Qwen3 8B (Non-reasoning)$0.25

Speed Comparison

Output Speed (tokens/s) — higher is better
Llama 3.2 Instruct 90B (Vision)
42 tok/s
Qwen3 8B (Non-reasoning)
81 tok/s
Time to First Token (seconds) — lower is better
Llama 3.2 Instruct 90B (Vision)
0.37s
Qwen3 8B (Non-reasoning)
0.95s

Benchmark Comparison

Data from Artificial Analysis API — 12 benchmarks

Intelligence Index
11.910.6
Coding Index
7.1
Math Index
24.3
GPQA Diamond
43.2%45.2%
MMLU-Pro
67.1%64.3%
LiveCodeBench
21.4%20.2%
AIME 2025
24.3%
MATH-500
62.9%82.8%
Humanity's Last Exam
4.9%2.8%
SciCode
24.0%16.8%
IFBench
28.6%
TerminalBench
2.3%
Llama 3.2 Instruct 90B (Vision)5 wins
7 winsQwen3 8B (Non-reasoning)

Frequently Asked Questions

Which is cheaper, Llama 3.2 Instruct 90B (Vision) or Qwen3 8B (Non-reasoning)?

Qwen3 8B (Non-reasoning) is cheaper overall. Its blended price (3:1 input/output ratio) is $0.31/M tokens vs $0.72/M for Llama 3.2 Instruct 90B (Vision).

Which model performs better on benchmarks?

Qwen3 8B (Non-reasoning) wins 7 out of 12 benchmarks compared to 5 for Llama 3.2 Instruct 90B (Vision). See the detailed benchmark chart above for per-category results.

Which is faster for real-time applications?

Qwen3 8B (Non-reasoning) generates tokens faster at 81 tok/s vs 42 tok/s. Llama 3.2 Instruct 90B (Vision) also has lower time-to-first-token (0.37s vs 0.95s).

When should I use Llama 3.2 Instruct 90B (Vision) vs Qwen3 8B (Non-reasoning)?

Choose based on your priorities: Qwen3 8B (Non-reasoning) for lower cost, Qwen3 8B (Non-reasoning) for stronger benchmark performance, and Qwen3 8B (Non-reasoning) for faster generation. For latency-sensitive apps, check the TTFT comparison above.