Model Comparator

Compare up to four models using the same criteria.

Apriel-v1.5-15B-Thinker
ServiceNow
DeepSeek V3.1 (Non-reasoning)
DeepSeek

Differences only

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

Model Comparator
Metrics	Apriel-v1.5-15B-Thinker ServiceNow	DeepSeek V3.1 (Non-reasoning) DeepSeek
Intelligence	Best: 13.8	13.7
Math	Best: 87.5	49.7
Speed	—	98 t/s
TTFT	—	1.39 s
Input · $ / 1M tokens	—	$0.56
Output · $ / 1M tokens	—	$1.68
Context window	128K	128K
Reasoning	Yes	No
Open weights	Yes	Yes

Intelligence

Apriel-v1.5-15B-Thinker: Best: 13.8
DeepSeek V3.1 (Non-reasoning): 13.7

Math

Apriel-v1.5-15B-Thinker: Best: 87.5
DeepSeek V3.1 (Non-reasoning): 49.7

Speed

Apriel-v1.5-15B-Thinker: —
DeepSeek V3.1 (Non-reasoning): 98 t/s

TTFT

Apriel-v1.5-15B-Thinker: —
DeepSeek V3.1 (Non-reasoning): 1.39 s

Input · $ / 1M tokens

Apriel-v1.5-15B-Thinker: —
DeepSeek V3.1 (Non-reasoning): $0.56

Output · $ / 1M tokens

Apriel-v1.5-15B-Thinker: —
DeepSeek V3.1 (Non-reasoning): $1.68

Context window

Apriel-v1.5-15B-Thinker: 128K
DeepSeek V3.1 (Non-reasoning): 128K

Reasoning

Apriel-v1.5-15B-Thinker: Yes
DeepSeek V3.1 (Non-reasoning): No

Open weights

Apriel-v1.5-15B-Thinker: Yes
DeepSeek V3.1 (Non-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.

DeepSeek V3.1 (Non-reasoning)13.7 · $0.84

Not plotted for lack of data on these axes: Apriel-v1.5-15B-Thinker.