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

Llama 4 Maverick
9.3
Gemini 2.0 Flash Thinking Experimental (Jan '25)
Best: 9.4

Coding

Llama 4 Maverick
16.3
Gemini 2.0 Flash Thinking Experimental (Jan '25)
Best: 24.1

Math

Llama 4 Maverick
19.3
Gemini 2.0 Flash Thinking Experimental (Jan '25)
—

Agentic

Llama 4 Maverick
0.6
Gemini 2.0 Flash Thinking Experimental (Jan '25)
—

Speed

Llama 4 Maverick
56.8 t/s
Gemini 2.0 Flash Thinking Experimental (Jan '25)
—

TTFT

Llama 4 Maverick
1.11 s
Gemini 2.0 Flash Thinking Experimental (Jan '25)
—

Input · $ / 1M tokens

Llama 4 Maverick
$0.25
Gemini 2.0 Flash Thinking Experimental (Jan '25)
—

Output · $ / 1M tokens

Llama 4 Maverick
$0.87
Gemini 2.0 Flash Thinking Experimental (Jan '25)
—

Context window

Llama 4 Maverick
1M
Gemini 2.0 Flash Thinking Experimental (Jan '25)
1M

Reasoning

Llama 4 Maverick
No
Gemini 2.0 Flash Thinking Experimental (Jan '25)
Yes

Open weights

Llama 4 Maverick
Yes
Gemini 2.0 Flash Thinking Experimental (Jan '25)
No

Cost / task

Llama 4 Maverick
$0.0382
Gemini 2.0 Flash Thinking Experimental (Jan '25)
—

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
  • Llama 4 Maverick9.3 · $0.0382

Not plotted for lack of data on these axes: Gemini 2.0 Flash Thinking Experimental (Jan '25).

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