Model

Gemini 1.0 Ultra

CLOSED

Google · Gemini 1.0 family · released Feb 8, 2024

Google's first frontier model; launched Gemini Advanced ($19.99/mo Google One AI Premium) on Feb 8 2024; now shut down (404s).

ReasoningCodingVisionFunction callingTool useAgentic

992.0

Elo · rank #214

Parameters

Undisclosed

Active params

—

Context

32K tokens

Architecture

Dense Transformer decoder with multi-query attention

License

Proprietary (Gemini Advanced / Vertex AI)

Languages

—

API price (in/out)

No hosted API

Modalities

text · vision · audio

Benchmark results

Bar shows position within the tracked field; marker = field best

AI2DVision79.5%#36

best: Molmo 72B · 96.3%

BIG-Bench HardReasoning83.6%#22

best: ERNIE 4.5 300B-A47B · 94.3%

ChartQAVision80.8%#34

best: MiniMax-VL-01 · 91.7%

DocVQAVision90.9%#29

best: Qwen2-VL-72B · 96.5%

DROPReasoning82.4#21

best: Hunyuan-T1 · 93.1

GSM8KMath94.4%#15

best: Llama 3.1 405B · 96.8%

HellaSwagReasoning87.8%#15

best: Claude 3 Opus · 95.4%

HumanEvalCoding74.4%#65

best: Claude Opus 4.5 · 99.4%

MATH-500Math53.2%#122

best: GPT-5 · 99.4%

MathVistaVision53.0%#50

best: Seed 2.1 Pro · 90.7%

MGSMMath79.0%#29

best: OpenAI o4-mini · 93.7%

MMLUKnowledge90.0%#9

best: OpenAI o3 · 92.9%

MMMUVision62.4%#73

best: Claude Fable 5 · 89.3%

TextVQAVision82.3%#9

best: Molmo 2 8B · 85.7%

VQAv2Vision77.8%#14

best: Reka Edge (2026) · 88.4%

Run it locally

Closed weights — available via API only. No local deployment.

Input / M tok

—

Output / M tok

—

Gemini 1.0 family

Elo progression across releases

Gemini 1.0 NanoDec 202391.9 Gemini 1.0 ProDec 2023620.1 Gemini 1.0 UltraFeb 2024992.0

Resources

DOCSstorage.googleapis.com/deepmind-media/gemini/gemini_1_report.pdf

API price — · each benchmark row carries its own source badge (see methodology)