Gemini 3.1 Pro

LLM Provider / Model · Long-horizon agentic tasks, multimodal reasoning and generative UI

At a glance

Input price	$2
Output price	$12
Cache price	$0.50
Price tier	Premium
Context window	1M+
Max output	64K
Context tier	500K+
Speed tier	Standard
Latency	medium
Modality	Multimodal (vision, audio, video)
Model ID	gemini-3.1-pro-preview
Provider	Google
Benchmark	Google’s current Pro tier; replaced the retired Gemini 3 Pro
Released	Feb 2026
Hosting	Closed/API
Capabilities	Vision, Audio, Video, Tool use, Deep Think, Generative UI, Streaming, Function calling, Prompt caching, Code execution

Vision, Audio, Video, Tool use, Deep Think, Generative UI, Streaming, Function calling, Prompt caching, Code execution

Long-horizon agentic tasks, multimodal reasoning and generative UI

Kiro's spec mode is tuned for Claude models — Gemini 3.1 Pro may work but spec-driven flows are best with Claude.
OpenAI Codex (VS Code) is tied to OpenAI's hosted GPT-5 and Codex models — Gemini 3.1 Pro is not selectable inside this integration.
Claude Code only works with Claude models — your LLM is Gemini 3.1 Pro.
Devin uses Cognition's proprietary model internally — your selected Gemini 3.1 Pro will be ignored.
OpenAI Codex (Cloud) runs GPT-5.1 Codex internally — your selected Gemini 3.1 Pro will be ignored.
The Claude Code SDK is tuned for Claude models — Gemini 3.1 Pro may work via the model adapter but loses Claude-specific features (thinking, prompt caching, sub-agents).
Lovable uses a fixed internal model — your selected Gemini 3.1 Pro will be ignored.
Replit Agent 3 uses its own model selection internally — your selected Gemini 3.1 Pro will be ignored.

Build a full stack around Gemini 3.1 Pro — Flowpicker shows compatibility warnings before you commit.