VERIFIED ORGANISATION

Google

Explore the documented models, published evaluations and primary sources associated with this organisation.

01 / MODELS

Verified catalogue

FAST MULTIMODAL

Gemini 3.8 Flash

Flash model for long-running software engineering, autonomous agents, and complex business flows.

16 SEP 2026 MULTIMODAL REASONING

Gemini 3.1 Pro

Multimodal reasoning, coding, and complex problems.

16 SEP 2026 PREVIOUS FLASH · AGENTS

Gemini 3.7 Flash

Previous Flash generation for complex code, agents, and multi-stage execution; useful as a comparison point with Gemini 3.8.

18 SEP 2026 EFFICIENT NATIVE IMAGE MODEL

Nano Banana 2

Production image model that combines generation, editing, and visual reasoning with a balance of quality, cost, and latency.

18 SEP 2026 CINEMATIC VIDEO WITH AUDIO

Veo 3.1

Cinematic video model with native audio, scene extension, and control through frames and reference images.

18 SEP 2026 MUSIC GENERATION

Lyria 3.5

Music model for full songs with vocals, lyrics, and control over structure, style, duration, and instrumentation.

18 SEP 2026 UNIFIED MULTIMODAL EMBEDDING

Gemini Embedding 2

Embedding model that places text, images, video, audio, and PDFs in a shared vector space for search and multimodal RAG.

18 SEP 2026 BUILT-IN SPATIAL REASONING

Gemini Robotics ER 2

Model specialized in physical understanding, video, spatial reasoning, and tool orchestration for robotic systems.

18 SEP 2026 REAL-TIME MULTIMODAL CONVERSATION

Gemini 3.8 Live

Live conversation model integrating voice, video, and tools with low latency.

29 SEP 2026 LIVE VOICE WITH EXTENDED REASONING

Gemini 3.8 Live Extended Thinking

Live voice variant with background reasoning for complex multi-step tasks.

29 SEP 2026
02 / EVALUATIONS

Published results

Figures are shown with the context reported by their source. A provider result is not an independent comparison and does not replace your own evaluation.

ModelBenchmarkResultMetric
Gemini 3.8 FlashArtificial Analysis Intelligence v4.150,2Index
Gemini 3.8 FlashGDPval-AA v21.348,8Elo
Gemini 3.1 ProSWE-Bench Pro54,2 %Resolved
Gemini 3.1 ProTerminal-Bench 2.170,7 %Accuracy
Gemini 3.7 FlashEvaluaciones Gemini 3.7 FlashPublishedCoding and agents
Nano Banana 2Evaluación visual Nano Banana 2Recommended modelQuality, cost, and latency
Veo 3.1MovieGenBenchPublished leadershipGlobal text-to-video preference
Lyria 3.5Evaluación de Lyria 3.5PublishedMusicality, vocals, and structural coherence
Gemini Embedding 2Evaluaciones Gemini Embedding 2PublishedMultimodal retrieval
Gemini Robotics ER 2Evaluaciones Robotics ER 2PublishedSpatial and robotic reasoning
03 / METHOD

How to read the directory

An organization, a product, and a model are not the same unit. Inferama separates them to avoid attributing capabilities or commercial terms to the wrong item.

01

Laboratory

Entity that develops or publishes the model and maintains its technical and safety documentation.

02

Model

Identifiable version with limits, modalities, and behavior that may change between releases.

03

Access channel

API, application, associated cloud, or commercial plan; each channel may have different pricing, retention, and limits.

04

Source and date

Every claim must retain the official page consulted and the verification date.

ANALYSIS

Related analysis

Gemini Robotics ER 2: What the Model Plans and What the Robot Must Validate
ANALISIS

Gemini Robotics ER 2: What the Model Plans and What the Robot Must Validate

Gemini Robotics ER 2 is presented as a vision-language model capable of planning multi-step tasks for robotics. That does not show that it directly controls a robot or that its plans are safe or reliable in every environment. We examine what the available documentation supports and what each team must test.

30 Sep 2026
Gemini Embedding 2: What We Know About Its Multimodal Embeddings—and What Remains to Be Evaluated
ANALISIS

Gemini Embedding 2: What We Know About Its Multimodal Embeddings—and What Remains to Be Evaluated

Google’s documentation describes a model that represents text and other types of content in a shared embedding space. That supports a stated capability, not a conclusion that it improves retrieval on any particular corpus. We review what the available sources can establish about performance, access, pricing, and data, and what teams should verify before testing it.

30 Sep 2026
Gemini 3.8 Flash: What We Know About Agents, Access, Pricing, and Safety
ANALISIS

Gemini 3.8 Flash: What We Know About Agents, Access, Pricing, and Safety

Google positions Gemini 3.8 Flash for long-horizon software engineering and autonomous agents. That stated focus does not, by itself, show that the model can reliably complete extended tasks. This review separates the provider’s claims from documented access and Agent Platform pricing conditions, and identifies evidence that is missing from the sources reviewed.

30 Sep 2026
Gemini 3.1 Pro with tools: how to test it before trusting it with a workflow
ANALISIS

Gemini 3.1 Pro with tools: how to test it before trusting it with a workflow

Google describes Gemini 3.1 Pro as a model for complex tasks, but that description does not prove that a specific integration can reliably complete a multi-step workflow. We propose a reversible evaluation protocol and explain what the available sources can—and cannot—establish about access, pricing, safety, and performance.

29 Sep 2026
Gemini 3.7 Flash: capabilities, pricing, and the limits of the evidence
ANALISIS

Gemini 3.7 Flash: capabilities, pricing, and the limits of the evidence

Google’s documentation presents Gemini 3.7 Flash as a model for multi-step tasks, code refactoring, and reasoning. Here is what the available sources can—and cannot—verify about access, pricing, safety, and performance.

29 Sep 2026
Gemini 3.1 Pro vs. Gemini 3.7 Flash for Code Maintenance: How to Measure Whether Paying More Pays Off
COMPARATIVA

Gemini 3.1 Pro vs. Gemini 3.7 Flash for Code Maintenance: How to Measure Whether Paying More Pays Off

There are no results from a controlled comparison here that would support declaring a winner. What we can offer is a reproducible protocol for measuring patch quality, regressions, cost, and time under shared conditions—and for deciding what evidence a team needs before choosing.

28 Sep 2026
04 / SOURCES

Traceability

TECHNICAL SHEETGemini 3.8 Flashai.google.dev↗PRICESGemini Developer API Pricingai.google.dev↗GUIDEGemini 3.8 Flash updatesai.google.dev↗BENCHMARKArtificial Analysis Intelligence v4.1openai.com↗PRIMARY SOURCEOfficial documentationai.google.dev↗TECHNICAL SHEETOfficial profile of Gemini 3.7 Flashai.google.dev↗CATALOGCatálogo oficial del proveedorai.google.dev↗TECHNICAL SHEETOfficial image generation guideai.google.dev↗TECHNICAL SHEETOfficial profile of Veodeepmind.google↗TECHNICAL SHEETOfficial profile of Lyriadeepmind.google↗CATALOGCatálogo oficial del proveedordeepmind.google↗TECHNICAL SHEETOfficial Gemini embeddings guideai.google.dev↗TECHNICAL SHEETFicha oficial del modeloai.google.dev↗TECHNICAL SHEETFicha oficial del modeloai.google.dev↗