Compare/Qwen3 4B (Reasoning) vs Gemma 4 E4B (Reasoning)

Qwen3 4B (Reasoning)vsGemma 4 E4B (Reasoning)

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

Alibaba

Qwen3 4B (Reasoning)

Input
$0.11/M
Output
$1.26/M
Speed
104 tok/s
TTFT
1.08s
Google

Gemma 4 E4B (Reasoning)

Input
$0/M
Output
$0/M
Speed
TTFT

Winner by Category

Cheaper
Gemma 4 E4B (Reasoning)
Faster (tok/s)
Qwen3 4B (Reasoning)
Lower Latency
Gemma 4 E4B (Reasoning)
Benchmarks (6-6)
Tie

Pricing Comparison

MetricQwen3 4B (Reasoning)Gemma 4 E4B (Reasoning)
Input ($/M tokens)$0.11$0
Output ($/M tokens)$1.26$0
Cost for 1M input + 100K output tokens:
Qwen3 4B (Reasoning)$0.24
Gemma 4 E4B (Reasoning)$0.00

Speed Comparison

Output Speed (tokens/s) — higher is better
Qwen3 4B (Reasoning)
104 tok/s
Gemma 4 E4B (Reasoning)
Time to First Token (seconds) — lower is better
Qwen3 4B (Reasoning)
1.08s
Gemma 4 E4B (Reasoning)

Benchmark Comparison

Data from Artificial Analysis API — 12 benchmarks

Intelligence Index
14.218.8
Coding Index
13.7
Math Index
22.3
GPQA Diamond
52.2%57.6%
MMLU-Pro
69.6%
LiveCodeBench
46.5%
AIME 2025
22.3%
MATH-500
93.3%
Humanity's Last Exam
5.1%3.7%
SciCode
3.5%24.4%
IFBench
32.5%44.2%
TerminalBench
8.3%
Qwen3 4B (Reasoning)6 wins
6 winsGemma 4 E4B (Reasoning)

Frequently Asked Questions

Which is cheaper, Qwen3 4B (Reasoning) or Gemma 4 E4B (Reasoning)?

Gemma 4 E4B (Reasoning) is cheaper overall. Its blended price (3:1 input/output ratio) is $0.00/M tokens vs $0.40/M for Qwen3 4B (Reasoning).

Which model performs better on benchmarks?

It's a tie — both models win 6 benchmarks each across 12 evaluated categories. See the detailed benchmark chart above for per-category results.

Which is faster for real-time applications?

Qwen3 4B (Reasoning) generates tokens faster at 104 tok/s vs 0 tok/s. However, Gemma 4 E4B (Reasoning) has lower time-to-first-token (0.00s vs 1.08s).

When should I use Qwen3 4B (Reasoning) vs Gemma 4 E4B (Reasoning)?

Choose based on your priorities: Gemma 4 E4B (Reasoning) for lower cost, both perform similarly on benchmarks, and Qwen3 4B (Reasoning) for faster generation. For latency-sensitive apps, check the TTFT comparison above.