Model Comparator

Compare up to four models using the same criteria.

Qwen3 Max (Preview)
Alibaba
Granite 4.2 30B
IBM

Differences only

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

Model Comparator
Metrics	Qwen3 Max (Preview) Alibaba	Granite 4.2 30B IBM
Intelligence	12.6	Best: 12.7
Coding	—	29.9
Math	75	—
Agentic	—	3.5
Speed	57.4 t/s	Best: 75.5 t/s
TTFT	3.96 s	Best: 0.87 s
Input · $ / 1M tokens	$1.20×7.5 relative to the lowest price among the compared models	Best: $0.16
Output · $ / 1M tokens	$6×9.2 relative to the lowest price among the compared models	Best: $0.65
Context window	Best: 262K	131K
Reasoning	No	Yes
Open weights	No	Yes
Cost / task	—	$0.0717

Intelligence

Qwen3 Max (Preview): 12.6
Granite 4.2 30B: Best: 12.7

Coding

Qwen3 Max (Preview): —
Granite 4.2 30B: 29.9

Math

Qwen3 Max (Preview): 75
Granite 4.2 30B: —

Agentic

Qwen3 Max (Preview): —
Granite 4.2 30B: 3.5

Speed

Qwen3 Max (Preview): 57.4 t/s
Granite 4.2 30B: Best: 75.5 t/s

TTFT

Qwen3 Max (Preview): 3.96 s
Granite 4.2 30B: Best: 0.87 s

Input · $ / 1M tokens

Qwen3 Max (Preview): $1.20×7.5 relative to the lowest price among the compared models
Granite 4.2 30B: Best: $0.16

Output · $ / 1M tokens

Qwen3 Max (Preview): $6×9.2 relative to the lowest price among the compared models
Granite 4.2 30B: Best: $0.65

Context window

Qwen3 Max (Preview): Best: 262K
Granite 4.2 30B: 131K

Reasoning

Qwen3 Max (Preview): No
Granite 4.2 30B: Yes

Open weights

Qwen3 Max (Preview): No
Granite 4.2 30B: Yes

Cost / task

Qwen3 Max (Preview): —
Granite 4.2 30B: $0.0717

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.

Qwen3 Max (Preview)12.6 · $2.40
Granite 4.2 30B12.7 · $0.282