Gemini

Google · United States · Language & multimodal
Global Closed sourceIndependent test
Showing version: Gemini 4 Carbon · 2026-10-28 · Version lineage ↓
Gemini 4 CarbonGemini 4 ArgonGemini 3.8 FlashGemini 3.5 FlashGemma 4 31BGemini 3.1 ProGemini 2.5 ProGemini 1.5 ProGemini 1.0
01

Overview

Google DeepMind's multimodal model family. Current flagships are Gemini 4 Carbon (restricted access) and Gemini 4 Argon; the open-weight Gemma line is recorded within this family, following how Google groups it on the official model pages.

Context length1,048,576 tokensInput$2.00 USD / 1M tokensRelease date2026-10-28
02 / Provenance

Field checks and source notes

When sources disagree on a field, every reading is listed here with its own citation. We do not take a midpoint, invent a missing figure, or decide which side is right — follow the citation and judge for your case.

These field-level notes are recorded verbatim in Chinese in the source data. Open the Chinese version

03 / Plain layer

Plain-language benchmarks

Smartness 60 · Top
Writing 1526 · Human preferencehigh 档;benchleader 复读同榜为 1527,存在 1 分读数差异
Memory 1,048,576 tokens
Model size Not disclosed
Price / free $2.00 / $10.00 USD / 1M tokens
04

Practical facts

Context length1,048,576 tokens
Max outputNot disclosed
ParametersNot disclosed
Input$2.00 USD / 1M tokens
Output price$10.00 USD / 1M tokens
CNYNot disclosed
Free tierNot disclosed
Output speedNot disclosed
Release date
LicenseNot disclosed
Show technical data

Technical benchmark table

This model is evaluated per reasoning tier; available tiers: low / medium / high. Each row below shows the highest tier measured for that benchmark. Tier switching happens on the compare page.

BenchmarkScoreMax / unit Source typeSourceTest dateConditions
AA Intelligence Index
aa-intelligence-index · v5
60 score_0_100 / 100 Independent test Tierhigh
AA Finance & Accounting
aa-finance-index · v5
62.6 score_0_100 / 100 Independent test Tierhigh
AA Strategy & Ops
aa-strategy-index · v5
60.9 score_0_100 / 100 Independent test Tierhigh
AA Legal
aa-legal-index · v5
62 score_0_100 / 100 Independent test Tierhigh
AA Economics
aa-economics-index · v5
59.6 score_0_100 / 100 Independent test Tierhigh
AA Engineering
aa-engineering-index · v5
65.3 score_0_100 / 100 Independent test Tierhigh
LMArena text preference
lmarena-text
1526 elo_like Human preference Tierhigh
05

Version lineage

Back to family page →
VersionRelease dateStatus OpennessParametersWhat changed Detail
Gemini 4 CarbonShowing version Active Closed source Not disclosed TBD Detail
Gemini 4 Argon Active Closed source Not disclosed TBD Detail
Gemini 3.8 Flash Active Closed source Not disclosed TBD Detail
Gemma 4 31B Active Open weights Not disclosed TBD Detail
Gemini 3.5 Flash Superseded Closed source Not disclosed TBD Detail
Gemini 3.1 Pro Superseded Closed source Not disclosed TBD Detail
Gemini 2.5 Pro Superseded Closed source Not disclosed TBD Detail
Gemini 1.5 Pro Discontinued Closed source Not disclosed TBD Detail
Gemini 1.0 Discontinued Closed source Not disclosed TBD Detail

Last updated: · Sources on detail page

06 / Compare

Comparable in this category

Baichuan
Claude
Command
DeepSeek
0 / 3 Compare Open comparison →