Model Comparator

Compare up to four models using the same criteria.

Mistral Saba
Mistral
DeepSeek R1 Distill Llama 8B
DeepSeek

Differences only

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

Model Comparator
Metrics	Mistral Saba Mistral	DeepSeek R1 Distill Llama 8B DeepSeek	Add a model · 2/4
Intelligence	6.5	6.5
Math	—	41.3
Context window	32K	Best: 128K
Reasoning	No	Yes
Open weights	No	Yes

Intelligence

Mistral Saba: 6.5
DeepSeek R1 Distill Llama 8B: 6.5

Math

Mistral Saba: —
DeepSeek R1 Distill Llama 8B: 41.3

Context window

Mistral Saba: 32K
DeepSeek R1 Distill Llama 8B: Best: 128K

Reasoning

Mistral Saba: No
DeepSeek R1 Distill Llama 8B: Yes

Open weights

Mistral Saba: No
DeepSeek R1 Distill Llama 8B: 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.