FR

Model Comparator

Compare up to four models using the same criteria.

Differences only

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

Model Comparator
Metrics	Qwen3 4B (Non-reasoning) Alibaba	DeepSeek-V2.5 DeepSeek	Add a model · 2/4
Intelligence	6.6	6.6
Context window	32K	Best: 128K
Reasoning	No	No
Open weights	Yes	Yes

Intelligence

Qwen3 4B (Non-reasoning): 6.6
DeepSeek-V2.5: 6.6

Context window

Qwen3 4B (Non-reasoning): 32K
DeepSeek-V2.5: Best: 128K

Reasoning

Qwen3 4B (Non-reasoning): No
DeepSeek-V2.5: No

Open weights

Qwen3 4B (Non-reasoning): Yes
DeepSeek-V2.5: Yes

Each reasoning level remains a separate configuration.

Preferences and analytics

Rybbit: visits, performance, clicks, copies, forms and session replay (masked inputs), with your permission. Refusing does not limit the site.