Compare / GPT-6.1 Sol vs Gemini 4 Argon

GPT-6.1 Sol vs Gemini 4 Argon

A side by side look at published price, context window and benchmark scores for these two models. Every number links to its source. A missing field means that model has not published a number for that metric.

What this means for your work

Both cost the same per token ($2 in, $10 out) and hold about 1.05 million tokens. Gemini 4 Argon is slightly higher on the Artificial Analysis index (53 vs 52) and produces text faster at the settings tested. GPT-6.1 Sol is higher on LMArena WebDev (1749 vs 1674). At the same price, choose by the job: app and site work leans Sol, quicker everyday answers lean Argon.

Input price

Lower is better

GPT-6.1 Sol $2.00

Gemini 4 Argon $2.00

GPT-6.1 Sol: source Vendor pricing page, as of Oct 27, 2026. standard text input per 1M tokens; prompts over 272K input tokens are 2x input for the full request

Gemini 4 Argon: source Vendor pricing page, as of Oct 27, 2026. paid standard per 1M tokens through 2026-11-30; $4.00 from 2026-12-01; page last updated 2026-10-27 UTC

Output price

Lower is better

GPT-6.1 Sol $10

Gemini 4 Argon $10

GPT-6.1 Sol: source Vendor pricing page, as of Oct 27, 2026. standard text output per 1M tokens; prompts over 272K input tokens are 1.5x output

Gemini 4 Argon: source Vendor pricing page, as of Oct 27, 2026. paid standard including thinking tokens through 2026-11-30; $20.00 from 2026-12-01

Context window

Higher is better

GPT-6.1 Sol 1.05M tokens

Gemini 4 Argon 1.05M tokens

GPT-6.1 Sol: source Vendor model page, as of Oct 27, 2026. 1,050,000 context window on the model details page

Gemini 4 Argon: source Vendor model page, as of Oct 14, 2026. unified input and output token limit; page last updated 2026-10-14 UTC

Output speed

Higher is better

GPT-6.1 Sol 65.9 tok/s

Gemini 4 Argon 80.9 tok/s

GPT-6.1 Sol: source Artificial Analysis, as of Oct 27, 2026. max; OpenAI API

Gemini 4 Argon: source Artificial Analysis, as of Oct 27, 2026. high; Google API; AA model-page output speed

Time to first token

Lower is better

For reasoning settings, Artificial Analysis counts thinking time before the first answer token, so long waits are expected on max effort modes.

GPT-6.1 Sol 325.58 s

Gemini 4 Argon 48.23 s

GPT-6.1 Sol: source Artificial Analysis, as of Oct 27, 2026. max; AA FAQ TTFT, includes thinking on this reasoning setting

Gemini 4 Argon: source Artificial Analysis, as of Oct 27, 2026. high; Google API; AA FAQ label is TTFT; on reasoning models AA sets that equal to time to first answer, so thinking is included

Artificial Analysis Intelligence Index

Higher is better

Composite of several evaluations run by Artificial Analysis at the setting shown.

GPT-6.1 Sol 52 points

Gemini 4 Argon 53 points

GPT-6.1 Sol: source Artificial Analysis, as of Oct 27, 2026. max effort; AA index integer

Gemini 4 Argon: source Artificial Analysis, as of Oct 27, 2026. high reasoning; AA Intelligence Index v4.3.2 integer shown on the model page

LMArena Text score

Higher is better

Bars start at 1300, not 0, to make differences easier to see.

GPT-6.1 Sol Not published

Gemini 4 Argon 1526

Gemini 4 Argon: source LMArena Text, as of Oct 22, 2026. Reported for Gemini 4 Argon (High).

LMArena WebDev score

Higher is better

Bars start at 1300, not 0, to make differences easier to see.

GPT-6.1 Sol 1749

Gemini 4 Argon 1674

GPT-6.1 Sol: source LMArena WebDev, as of Oct 23, 2026. Reported for GPT-6.1 Sol (Max).

Gemini 4 Argon: source LMArena WebDev, as of Oct 23, 2026. Reported for Gemini 4 Argon (High).

All metrics

Metric	GPT-6.1 Sol	Gemini 4 Argon
Input price	$2.00as of Oct 27, 2026	$2.00as of Oct 27, 2026
Output price	$10as of Oct 27, 2026	$10as of Oct 27, 2026
Context window	1.05M tokensas of Oct 27, 2026	1.05M tokensas of Oct 14, 2026
Output speed	65.9 tok/sas of Oct 27, 2026	80.9 tok/sas of Oct 27, 2026
Time to first token	325.58 sas of Oct 27, 2026	48.23 sas of Oct 27, 2026
Artificial Analysis Intelligence Index	52 pointsas of Oct 27, 2026	53 pointsas of Oct 27, 2026
LMArena Text score	Not published	1526as of Oct 22, 2026
LMArena WebDev score	1749as of Oct 23, 2026	1674as of Oct 23, 2026

Compare other models