Model Comparator

Compare up to four models using the same criteria.

DeepSeek V3 (Dec '24)
DeepSeek
DeepSeek R1 Distill Qwen 32B
DeepSeek

Differences only

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

Model Comparator
Metrics	DeepSeek V3 (Dec '24) DeepSeek	DeepSeek R1 Distill Qwen 32B DeepSeek
Intelligence	Best: 8.5	8.4
Coding	23	—
Math	26	Best: 63
Agentic	0.8	—
Speed	37 t/s	—
TTFT	1.4 s	—
Input · $ / 1M tokens	$0.32	—
Output · $ / 1M tokens	$0.89	—
Context window	128K	128K
Reasoning	No	Yes
Open weights	Yes	Yes
Cost / task	$0.0198	—

Intelligence

DeepSeek V3 (Dec '24): Best: 8.5
DeepSeek R1 Distill Qwen 32B: 8.4

Coding

DeepSeek V3 (Dec '24): 23
DeepSeek R1 Distill Qwen 32B: —

Math

DeepSeek V3 (Dec '24): 26
DeepSeek R1 Distill Qwen 32B: Best: 63

Agentic

DeepSeek V3 (Dec '24): 0.8
DeepSeek R1 Distill Qwen 32B: —

Speed

DeepSeek V3 (Dec '24): 37 t/s
DeepSeek R1 Distill Qwen 32B: —

TTFT

DeepSeek V3 (Dec '24): 1.4 s
DeepSeek R1 Distill Qwen 32B: —

Input · $ / 1M tokens

DeepSeek V3 (Dec '24): $0.32
DeepSeek R1 Distill Qwen 32B: —

Output · $ / 1M tokens

DeepSeek V3 (Dec '24): $0.89
DeepSeek R1 Distill Qwen 32B: —

Context window

DeepSeek V3 (Dec '24): 128K
DeepSeek R1 Distill Qwen 32B: 128K

Reasoning

DeepSeek V3 (Dec '24): No
DeepSeek R1 Distill Qwen 32B: Yes

Open weights

DeepSeek V3 (Dec '24): Yes
DeepSeek R1 Distill Qwen 32B: Yes

Cost / task

DeepSeek V3 (Dec '24): $0.0198
DeepSeek R1 Distill Qwen 32B: —

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 (Dec '24)8.5 · $0.0198

Not plotted for lack of data on these axes: DeepSeek R1 Distill Qwen 32B.