Model Comparator

Compare up to four models using the same criteria.

Gemma 4 26B A4B (Non-reasoning)
Google
EXAONE 4.5 33B (Reasoning)
LG AI Research

Differences only

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

Model Comparator
Metrics	Gemma 4 26B A4B (Non-reasoning) Google	EXAONE 4.5 33B (Reasoning) LG AI Research
Intelligence	13.1	Best: 13.2
Coding	—	23.6
Speed	93.5 t/s	—
TTFT	0.98 s	—
Input · $ / 1M tokens	$0.13	—
Output · $ / 1M tokens	$0.40	—
Context window	256K	Best: 262K
Reasoning	No	Yes
Open weights	Yes	Yes

Intelligence

Gemma 4 26B A4B (Non-reasoning): 13.1
EXAONE 4.5 33B (Reasoning): Best: 13.2

Coding

Gemma 4 26B A4B (Non-reasoning): —
EXAONE 4.5 33B (Reasoning): 23.6

Speed

Gemma 4 26B A4B (Non-reasoning): 93.5 t/s
EXAONE 4.5 33B (Reasoning): —

TTFT

Gemma 4 26B A4B (Non-reasoning): 0.98 s
EXAONE 4.5 33B (Reasoning): —

Input · $ / 1M tokens

Gemma 4 26B A4B (Non-reasoning): $0.13
EXAONE 4.5 33B (Reasoning): —

Output · $ / 1M tokens

Gemma 4 26B A4B (Non-reasoning): $0.40
EXAONE 4.5 33B (Reasoning): —

Context window

Gemma 4 26B A4B (Non-reasoning): 256K
EXAONE 4.5 33B (Reasoning): Best: 262K

Reasoning

Gemma 4 26B A4B (Non-reasoning): No
EXAONE 4.5 33B (Reasoning): Yes

Open weights

Gemma 4 26B A4B (Non-reasoning): Yes
EXAONE 4.5 33B (Reasoning): 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.

Gemma 4 26B A4B (Non-reasoning)13.1 · $0.198

Not plotted for lack of data on these axes: EXAONE 4.5 33B (Reasoning).