Model Comparator

Compare up to four models using the same criteria.

Mistral Small 4 (Reasoning)
Mistral
Granite 4.0 350M
IBM

Differences only

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

Model Comparator
Metrics	Mistral Small 4 (Reasoning) Mistral	Granite 4.0 350M IBM
Intelligence	4.3	Best: 4.8
Coding	26.6	—
Math	—	0
Agentic	0.8	—
Speed	166 t/s	—
TTFT	0.92 s	—
Input · $ / 1M tokens	$0.15	—
Output · $ / 1M tokens	$0.60	—
Context window	Best: 256K	33K
Reasoning	Yes	No
Open weights	Yes	Yes
Cost / task	$0.0399	—

Intelligence

Mistral Small 4 (Reasoning): 4.3
Granite 4.0 350M: Best: 4.8

Coding

Mistral Small 4 (Reasoning): 26.6
Granite 4.0 350M: —

Math

Mistral Small 4 (Reasoning): —
Granite 4.0 350M: 0

Agentic

Mistral Small 4 (Reasoning): 0.8
Granite 4.0 350M: —

Speed

Mistral Small 4 (Reasoning): 166 t/s
Granite 4.0 350M: —

TTFT

Mistral Small 4 (Reasoning): 0.92 s
Granite 4.0 350M: —

Input · $ / 1M tokens

Mistral Small 4 (Reasoning): $0.15
Granite 4.0 350M: —

Output · $ / 1M tokens

Mistral Small 4 (Reasoning): $0.60
Granite 4.0 350M: —

Context window

Mistral Small 4 (Reasoning): Best: 256K
Granite 4.0 350M: 33K

Reasoning

Mistral Small 4 (Reasoning): Yes
Granite 4.0 350M: No

Open weights

Mistral Small 4 (Reasoning): Yes
Granite 4.0 350M: Yes

Cost / task

Mistral Small 4 (Reasoning): $0.0399
Granite 4.0 350M: —

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 4 (Reasoning)4.3 · $0.0399

Not plotted for lack of data on these axes: Granite 4.0 350M.