Model Comparator

Compare up to four models using the same criteria.

Mistral Small (Sep '24)
Mistral
Phi-3 Mini Instruct 3.8B
Microsoft

Differences only

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

Model Comparator
Metrics	Mistral Small (Sep '24) Mistral	Phi-3 Mini Instruct 3.8B Microsoft
Intelligence	5.8	5.8
Math	—	0.3
Speed	154 t/s	—
TTFT	0.76 s	—
Input · $ / 1M tokens	$0.20	—
Output · $ / 1M tokens	$0.60	—
Context window	Best: 33K	4K
Reasoning	No	No
Open weights	Yes	Yes

Intelligence

Mistral Small (Sep '24): 5.8
Phi-3 Mini Instruct 3.8B: 5.8

Math

Mistral Small (Sep '24): —
Phi-3 Mini Instruct 3.8B: 0.3

Speed

Mistral Small (Sep '24): 154 t/s
Phi-3 Mini Instruct 3.8B: —

TTFT

Mistral Small (Sep '24): 0.76 s
Phi-3 Mini Instruct 3.8B: —

Input · $ / 1M tokens

Mistral Small (Sep '24): $0.20
Phi-3 Mini Instruct 3.8B: —

Output · $ / 1M tokens

Mistral Small (Sep '24): $0.60
Phi-3 Mini Instruct 3.8B: —

Context window

Mistral Small (Sep '24): Best: 33K
Phi-3 Mini Instruct 3.8B: 4K

Reasoning

Mistral Small (Sep '24): No
Phi-3 Mini Instruct 3.8B: No

Open weights

Mistral Small (Sep '24): Yes
Phi-3 Mini Instruct 3.8B: 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.

Mistral Small (Sep '24)5.8 · $0.30

Not plotted for lack of data on these axes: Phi-3 Mini Instruct 3.8B.