Gemini 4 Argon (Medium) vs GPT-5.6 Sol (Max)

Gemini 4 Argon (Medium) scores 47.6 on the Artificial Analysis Intelligence Index, against 47.0 for GPT-5.6 Sol (Max). GPT-5.6 Sol (Max) generates 108 t/s, against 64.5 t/s for Gemini 4 Argon (Medium). Both accept a context of 1M tokens.

Comparison generated from the catalogue data. Add or remove models below to refine it.

2 models compared

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

Intelligence

Gemini 4 Argon (Medium)
Best: 47.6
GPT-5.6 Sol (Max)
47

Coding

Gemini 4 Argon (Medium)
—
GPT-5.6 Sol (Max)
78.3

Agentic

Gemini 4 Argon (Medium)
—
GPT-5.6 Sol (Max)
47.4

Speed

Gemini 4 Argon (Medium)
64.5 t/s
GPT-5.6 Sol (Max)
Best: 108 t/s

TTFT

Gemini 4 Argon (Medium)
Best: 11.6 s
GPT-5.6 Sol (Max)
33.8 s

Input · $ / 1M tokens

Gemini 4 Argon (Medium)
$4
GPT-5.6 Sol (Max)
$4

Output · $ / 1M tokens

Gemini 4 Argon (Medium)
$20
GPT-5.6 Sol (Max)
$20

Context window

Gemini 4 Argon (Medium)
1M
GPT-5.6 Sol (Max)
1M

Reasoning

Gemini 4 Argon (Medium)
Yes
GPT-5.6 Sol (Max)
Yes

Open weights

Gemini 4 Argon (Medium)
No
GPT-5.6 Sol (Max)
No

Cost / task

Gemini 4 Argon (Medium)
$8.63×4.3 relative to the lowest price among the compared models
GPT-5.6 Sol (Max)
Best: $1.99

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
  • Gemini 4 Argon (Medium)47.6 · $8.63
  • GPT-5.6 Sol (Max)47 · $1.99

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