Model Comparator

Compare up to four models using the same criteria.

DeepSeek V3.1 (Reasoning)
DeepSeek
Nemotron 3.5 Super VL
NVIDIA

Differences only

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

Model Comparator
Metrics	DeepSeek V3.1 (Reasoning) DeepSeek	Nemotron 3.5 Super VL NVIDIA
Intelligence	13.5	Best: 13.6
Math	89.7	—
Speed	123 t/s	Best: 415 t/s
TTFT	2.06 s	Best: 0.71 s
Input · $ / 1M tokens	$0.58×2.9 relative to the lowest price among the compared models	Best: $0.20
Output · $ / 1M tokens	$1.68×3.4 relative to the lowest price among the compared models	Best: $0.50
Context window	1M	1M
Reasoning	Yes	Yes
Open weights	Yes	Yes
Cost / task	—	$0.538

Intelligence

DeepSeek V3.1 (Reasoning): 13.5
Nemotron 3.5 Super VL: Best: 13.6

Math

DeepSeek V3.1 (Reasoning): 89.7
Nemotron 3.5 Super VL: —

Speed

DeepSeek V3.1 (Reasoning): 123 t/s
Nemotron 3.5 Super VL: Best: 415 t/s

TTFT

DeepSeek V3.1 (Reasoning): 2.06 s
Nemotron 3.5 Super VL: Best: 0.71 s

Input · $ / 1M tokens

DeepSeek V3.1 (Reasoning): $0.58×2.9 relative to the lowest price among the compared models
Nemotron 3.5 Super VL: Best: $0.20

Output · $ / 1M tokens

DeepSeek V3.1 (Reasoning): $1.68×3.4 relative to the lowest price among the compared models
Nemotron 3.5 Super VL: Best: $0.50

Context window

DeepSeek V3.1 (Reasoning): 1M
Nemotron 3.5 Super VL: 1M

Reasoning

DeepSeek V3.1 (Reasoning): Yes
Nemotron 3.5 Super VL: Yes

Open weights

DeepSeek V3.1 (Reasoning): Yes
Nemotron 3.5 Super VL: Yes

Cost / task

DeepSeek V3.1 (Reasoning): —
Nemotron 3.5 Super VL: $0.538

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 (Reasoning)13.5 · $0.855
Nemotron 3.5 Super VL13.6 · $0.275