Model Comparator

Compare up to four models using the same criteria.

GPT-5.4 mini (Non-reasoning)
OpenAI
Granite 4.2 8B
IBM

Differences only

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

Model Comparator
Metrics	GPT-5.4 mini (Non-reasoning) OpenAI	Granite 4.2 8B IBM
Intelligence	11.1	11.1
Coding	—	22.4
Agentic	—	1.3
Speed	Best: 183 t/s	118 t/s
TTFT	Best: 0.67 s	0.85 s
Input · $ / 1M tokens	$0.75×12.5 relative to the lowest price among the compared models	Best: $0.060
Output · $ / 1M tokens	$4.50×18 relative to the lowest price among the compared models	Best: $0.25
Context window	Best: 400K	131K
Reasoning	No	Yes
Open weights	No	Yes
Cost / task	—	$0.0236

Intelligence

GPT-5.4 mini (Non-reasoning): 11.1
Granite 4.2 8B: 11.1

Coding

GPT-5.4 mini (Non-reasoning): —
Granite 4.2 8B: 22.4

Agentic

GPT-5.4 mini (Non-reasoning): —
Granite 4.2 8B: 1.3

Speed

GPT-5.4 mini (Non-reasoning): Best: 183 t/s
Granite 4.2 8B: 118 t/s

TTFT

GPT-5.4 mini (Non-reasoning): Best: 0.67 s
Granite 4.2 8B: 0.85 s

Input · $ / 1M tokens

GPT-5.4 mini (Non-reasoning): $0.75×12.5 relative to the lowest price among the compared models
Granite 4.2 8B: Best: $0.060

Output · $ / 1M tokens

GPT-5.4 mini (Non-reasoning): $4.50×18 relative to the lowest price among the compared models
Granite 4.2 8B: Best: $0.25

Context window

GPT-5.4 mini (Non-reasoning): Best: 400K
Granite 4.2 8B: 131K

Reasoning

GPT-5.4 mini (Non-reasoning): No
Granite 4.2 8B: Yes

Open weights

GPT-5.4 mini (Non-reasoning): No
Granite 4.2 8B: Yes

Cost / task

GPT-5.4 mini (Non-reasoning): —
Granite 4.2 8B: $0.0236

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.

GPT-5.4 mini (Non-reasoning)11.1 · $1.69
Granite 4.2 8B11.1 · $0.107