Model Comparator

Compare up to four models using the same criteria.

Qwen2.5 Instruct 72B
Alibaba
Llama 3.3 Instruct 70B
Meta

Differences only

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

Model Comparator
Metrics	Qwen2.5 Instruct 72B Alibaba	Llama 3.3 Instruct 70B Meta
Intelligence	7.7	7.7
Coding	—	11.9
Math	Best: 14	7.7
Speed	30.9 t/s	Best: 92.7 t/s
TTFT	3.92 s	Best: 1.71 s
Input · $ / 1M tokens	Best: $0.475	$0.71×1.5 relative to the lowest price among the compared models
Output · $ / 1M tokens	Best: $0.495	$0.72×1.5 relative to the lowest price among the compared models
Context window	Best: 131K	128K
Reasoning	No	No
Open weights	Yes	Yes

Intelligence

Qwen2.5 Instruct 72B: 7.7
Llama 3.3 Instruct 70B: 7.7

Coding

Qwen2.5 Instruct 72B: —
Llama 3.3 Instruct 70B: 11.9

Math

Qwen2.5 Instruct 72B: Best: 14
Llama 3.3 Instruct 70B: 7.7

Speed

Qwen2.5 Instruct 72B: 30.9 t/s
Llama 3.3 Instruct 70B: Best: 92.7 t/s

TTFT

Qwen2.5 Instruct 72B: 3.92 s
Llama 3.3 Instruct 70B: Best: 1.71 s

Input · $ / 1M tokens

Qwen2.5 Instruct 72B: Best: $0.475
Llama 3.3 Instruct 70B: $0.71×1.5 relative to the lowest price among the compared models

Output · $ / 1M tokens

Qwen2.5 Instruct 72B: Best: $0.495
Llama 3.3 Instruct 70B: $0.72×1.5 relative to the lowest price among the compared models

Context window

Qwen2.5 Instruct 72B: Best: 131K
Llama 3.3 Instruct 70B: 128K

Reasoning

Qwen2.5 Instruct 72B: No
Llama 3.3 Instruct 70B: No

Open weights

Qwen2.5 Instruct 72B: Yes
Llama 3.3 Instruct 70B: 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.

Qwen2.5 Instruct 72B7.7 · $0.48
Llama 3.3 Instruct 70B7.7 · $0.712