Model Comparator

Compare up to four models using the same criteria.

Claude 4 Sonnet (Non-reasoning)
Anthropic
Gemma 4 26B A4B (Reasoning)
Google

Differences only

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

Model Comparator
Metrics	Claude 4 Sonnet (Non-reasoning) Anthropic	Gemma 4 26B A4B (Reasoning) Google
Intelligence	16.6	Best: 16.7
Coding	37.6	Best: 39.3
Math	38	—
Agentic	—	2.9
Speed	—	65.1 t/s
TTFT	—	0.97 s
Input · $ / 1M tokens	—	$0.10
Output · $ / 1M tokens	—	$0.40
Context window	Best: 1M	256K
Reasoning	No	Yes
Open weights	No	Yes

Intelligence

Claude 4 Sonnet (Non-reasoning): 16.6
Gemma 4 26B A4B (Reasoning): Best: 16.7

Coding

Claude 4 Sonnet (Non-reasoning): 37.6
Gemma 4 26B A4B (Reasoning): Best: 39.3

Math

Claude 4 Sonnet (Non-reasoning): 38
Gemma 4 26B A4B (Reasoning): —

Agentic

Claude 4 Sonnet (Non-reasoning): —
Gemma 4 26B A4B (Reasoning): 2.9

Speed

Claude 4 Sonnet (Non-reasoning): —
Gemma 4 26B A4B (Reasoning): 65.1 t/s

TTFT

Claude 4 Sonnet (Non-reasoning): —
Gemma 4 26B A4B (Reasoning): 0.97 s

Input · $ / 1M tokens

Claude 4 Sonnet (Non-reasoning): —
Gemma 4 26B A4B (Reasoning): $0.10

Output · $ / 1M tokens

Claude 4 Sonnet (Non-reasoning): —
Gemma 4 26B A4B (Reasoning): $0.40

Context window

Claude 4 Sonnet (Non-reasoning): Best: 1M
Gemma 4 26B A4B (Reasoning): 256K

Reasoning

Claude 4 Sonnet (Non-reasoning): No
Gemma 4 26B A4B (Reasoning): Yes

Open weights

Claude 4 Sonnet (Non-reasoning): No
Gemma 4 26B A4B (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 (Reasoning)16.7 · $0.175

Not plotted for lack of data on these axes: Claude 4 Sonnet (Non-reasoning).