Model Comparator

Compare up to four models using the same criteria.

Gemma 4 12B (Non-reasoning)
Google
Qwen3 VL 30B A3B (Reasoning)
Alibaba

Differences only

✓ Best published value among measured models. — Data unavailable. These metrics do not establish an overall ranking.

Model Comparator
Metrics	Gemma 4 12B (Non-reasoning) Google	Qwen3 VL 30B A3B (Reasoning) Alibaba
Intelligence	9.4	Best: 9.5
Math	—	82.3
Speed	111 t/s	Best: 114 t/s
TTFT	2.49 s	Best: 2.25 s
Input · $ / 1M tokens	Best: $0.10	$0.50×5 relative to the lowest price among the compared models
Output · $ / 1M tokens	Best: $0.30	$0.50×1.7 relative to the lowest price among the compared models
Context window	Best: 262K	256K
Reasoning	No	Yes
Open weights	Yes	Yes

Intelligence

Gemma 4 12B (Non-reasoning): 9.4
Qwen3 VL 30B A3B (Reasoning): Best: 9.5

Math

Gemma 4 12B (Non-reasoning): —
Qwen3 VL 30B A3B (Reasoning): 82.3

Speed

Gemma 4 12B (Non-reasoning): 111 t/s
Qwen3 VL 30B A3B (Reasoning): Best: 114 t/s

TTFT

Gemma 4 12B (Non-reasoning): 2.49 s
Qwen3 VL 30B A3B (Reasoning): Best: 2.25 s

Input · $ / 1M tokens

Gemma 4 12B (Non-reasoning): Best: $0.10
Qwen3 VL 30B A3B (Reasoning): $0.50×5 relative to the lowest price among the compared models

Output · $ / 1M tokens

Gemma 4 12B (Non-reasoning): Best: $0.30
Qwen3 VL 30B A3B (Reasoning): $0.50×1.7 relative to the lowest price among the compared models

Context window

Gemma 4 12B (Non-reasoning): Best: 262K
Qwen3 VL 30B A3B (Reasoning): 256K

Reasoning

Gemma 4 12B (Non-reasoning): No
Qwen3 VL 30B A3B (Reasoning): Yes

Open weights

Gemma 4 12B (Non-reasoning): Yes
Qwen3 VL 30B A3B (Reasoning): Yes

Each reasoning level remains a separate configuration.

Score and cost

Each line joins the reasoning levels of a compared model; the large point marks the configuration in the table. The top-right corner holds the best score for the lowest cost. Select another point to compare that level.

Gemma 4 12B (Non-reasoning)9.4 · $0.15
Qwen3 VL 30B A3B (Reasoning)9.5 · $0.50