Neural Audio Synthesis
Pro
Gemini 3.1 Flash TTS is a neural speech synthesis model by Google, providing expressive voice rendering for real-time conversation.
128K ctx
Specialized Modality
The AnalystsStrict ZDR Verified
Large-context analysis and data processing.
Laboratory Overview
Google DeepMind delivers frontier multimodality and extreme long-context retrieval through the Gemini family. Offering up to 2M token context windows, native audio TTS, and Nano Banana visual models, Google models excel in research and media.
Core Research Focus:Large context analysisdata processingabstract reasoningagentic tool orchestrationscientific contentfrontier-tier multimodal understandinghigh-speed agentic executionautonomous software engineering and vulnerability discovery (3.8 Flash).
Privacy & Zero Data Retention
Strict Zero Data Retention (ZDR): Requests are processed ephemerally in RAM. No user data, prompts, or completions are stored, indexed, or used for model training.
No user telemetry sent to Google
Google Foundation Models (14)
Available under ARMES unified subscription without separate API keys or billing agreements.
Tier:
Dense Transformer
Pro
Abstract reasoning puzzles, competitive algorithms, agentic tool orchestration, scientific coding, large document analysis (10K+ tokens), technical/academic writing
ARC-AGI-2:77.1%
1M ctx
$2.00 / $12.00 /MTok
1M ctx
$1.25 / $10.00 /MTok
Mixture of Experts (MoE)
MoEEco
Google's next-generation workhorse model (released Sep 2, 2026; codenamed "Skimaki").
Terminal-Bench 2.1:90.8%
1M ctx
$1.50 / $7.50 /MTok
Dense Transformer
Pro
Google's former workhorse model (released Aug 13, 2026; knowledge cutoff Mar 2026).
Artificial Analysis Intelligence Index:56
1M ctx
$0.75 / $3.75 /MTok
Dense Transformer
Pro
Gemini 3.6 Flash is a high-performance foundation model engineered by Google, accessible with Zero Data Retention on ARMES.
128K ctx
$1.50 / $7.50 /MTok
Mixture of Experts (MoE)
MoEEco
Sustained frontier-level intelligence for agentic execution, iterative coding cycles, multi-step tool use, finance analysis, and professional knowledge work — at Flash speed.
Terminal-Bench 2.1:76.2%
1M ctx
$1.50 / $9.00 /MTok
Mixture of Experts (MoE)
MoEEco
Mathematics, science, PhD-level reasoning, data analysis, financial analysis, technical docs, competitive algorithms.
GPQA Diamond:90.4%
1M ctx
$0.50 / $3.00 /MTok
1M ctx
$0.30 / $2.50 /MTok
Mixture of Experts (MoE)
MoEEco
Gemini 3.5 Flash-Lite is a high-performance foundation model engineered by Google, accessible with Zero Data Retention on ARMES.
128K ctx
$0.30 / $2.50 /MTok
Dense Transformer
Free
Mathematics, science, data analysis, financial reasoning, tech docs, academic writing, logic puzzles, document summarization.
GPQA Diamond:86.9%
1M ctx
$0.25 / $1.50 /MTok
Diffusion Vision Synthesis
Pro
Nano Banana 2 Lite is an advanced image synthesis engine by Google, offering high-fidelity photorealistic rendering and prompt adherence.
128K ctx
Specialized Modality
Diffusion Vision Synthesis
Pro
Nano Banana 2 Flash is an advanced image synthesis engine by Google, offering high-fidelity photorealistic rendering and prompt adherence.
128K ctx
Specialized Modality
Diffusion Vision Synthesis
Pro
Nano Banana 2 Pro is an advanced image synthesis engine by Google, offering high-fidelity photorealistic rendering and prompt adherence.
128K ctx
Specialized Modality
Explore Other AI Labs on ARMES
Compare models across the 21+ leading laboratories unified in your workspace.
