Compare/Qwen3 8B (Non-reasoning) vs Mercury 2

Qwen3 8B (Non-reasoning)vsMercury 2

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

Alibaba

Qwen3 8B (Non-reasoning)

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

Mercury 2

Input
$0.25/M
Output
$0.75/M
Speed
907 tok/s
TTFT
3.76s

Winner by Category

Cheaper
Qwen3 8B (Non-reasoning)
Faster (tok/s)
Mercury 2
Lower Latency
Qwen3 8B (Non-reasoning)
Benchmarks (5-7)
Mercury 2

Pricing Comparison

MetricQwen3 8B (Non-reasoning)Mercury 2
Input ($/M tokens)$0.18$0.25
Output ($/M tokens)$0.7$0.75
Cost for 1M input + 100K output tokens:
Qwen3 8B (Non-reasoning)$0.25
Mercury 2$0.33

Speed Comparison

Output Speed (tokens/s) — higher is better
Qwen3 8B (Non-reasoning)
81 tok/s
Mercury 2
907 tok/s
Time to First Token (seconds) — lower is better
Qwen3 8B (Non-reasoning)
0.95s
Mercury 2
3.76s

Benchmark Comparison

Data from Artificial Analysis API — 12 benchmarks

Intelligence Index
10.632.8
Coding Index
7.130.6
Math Index
24.3
GPQA Diamond
45.2%77.0%
MMLU-Pro
64.3%
LiveCodeBench
20.2%
AIME 2025
24.3%
MATH-500
82.8%
Humanity's Last Exam
2.8%15.5%
SciCode
16.8%38.7%
IFBench
28.6%69.8%
TerminalBench
2.3%26.5%
Qwen3 8B (Non-reasoning)5 wins
7 winsMercury 2

Frequently Asked Questions

Which is cheaper, Qwen3 8B (Non-reasoning) or Mercury 2?

Qwen3 8B (Non-reasoning) is cheaper overall. Its blended price (3:1 input/output ratio) is $0.31/M tokens vs $0.38/M for Mercury 2.

Which model performs better on benchmarks?

Mercury 2 wins 7 out of 12 benchmarks compared to 5 for Qwen3 8B (Non-reasoning). See the detailed benchmark chart above for per-category results.

Which is faster for real-time applications?

Mercury 2 generates tokens faster at 907 tok/s vs 81 tok/s. Qwen3 8B (Non-reasoning) also has lower time-to-first-token (0.95s vs 3.76s).

When should I use Qwen3 8B (Non-reasoning) vs Mercury 2?

Choose based on your priorities: Qwen3 8B (Non-reasoning) for lower cost, Mercury 2 for stronger benchmark performance, and Mercury 2 for faster generation. For latency-sensitive apps, check the TTFT comparison above.