Model Comparator

Compare up to four models using the same criteria.

2 models compared

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

Intelligence

Magistral Medium 1.2
11.8
Nemotron 3 Ultra 550B A55B (Reasoning)
11.8

Coding

Magistral Medium 1.2
21.3
Nemotron 3 Ultra 550B A55B (Reasoning)
Best: 49.3

Math

Magistral Medium 1.2
82
Nemotron 3 Ultra 550B A55B (Reasoning)
—

Agentic

Magistral Medium 1.2
—
Nemotron 3 Ultra 550B A55B (Reasoning)
20.1

Cyber Index

Magistral Medium 1.2
—
Nemotron 3 Ultra 550B A55B (Reasoning)
13.1%

Speed

Magistral Medium 1.2
—
Nemotron 3 Ultra 550B A55B (Reasoning)
159 t/s

TTFT

Magistral Medium 1.2
—
Nemotron 3 Ultra 550B A55B (Reasoning)
1.23 s

Input · $ / 1M tokens

Magistral Medium 1.2
—
Nemotron 3 Ultra 550B A55B (Reasoning)
$0.60

Output · $ / 1M tokens

Magistral Medium 1.2
—
Nemotron 3 Ultra 550B A55B (Reasoning)
$2.70

Context window

Magistral Medium 1.2
128K
Nemotron 3 Ultra 550B A55B (Reasoning)
Best: 262K

Reasoning

Magistral Medium 1.2
Yes
Nemotron 3 Ultra 550B A55B (Reasoning)
Yes

Open weights

Magistral Medium 1.2
No
Nemotron 3 Ultra 550B A55B (Reasoning)
Yes

Cost / task

Magistral Medium 1.2
—
Nemotron 3 Ultra 550B A55B (Reasoning)
$2.23

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.

Intelligence Indexmost efficient ↗Cost per task
  • Nemotron 3 Ultra 550B A55B (Reasoning)11.8 · $2.23

Not plotted for lack of data on these axes: Magistral Medium 1.2.

Preferences and analytics

Rybbit: visits, performance, clicks, copies, forms and session replay (masked inputs), with your permission. Refusing does not limit the site.

Privacy

Allow navigation analytics?

With your permission, Rybbit measures visits, performance and interactions (clicks, copies, forms), including session replay, to improve BenchSift. Input values are masked in replays. Refusing does not limit the site.

Your choice is stored for 180 days. Change it at any time using “Privacy preferences” in the footer. Withdrawal reloads the page.

No choice stored: Rybbit is disabled.

Privacy