Model details
Gemini 3.1 Pro Preview Custom Tools is a specialized variant of Google's flagship language model that takes a more disciplined approach to tool selection. Rather than defaulting to a generic bash tool when third-party or user-defined functions are available, this model variant has been tuned to recognize and select the most efficient tool for the task at hand. This refinement is particularly valuable for coding agents and complex multi-tool workflows where the difference between the right tool and the default tool can significantly impact reliability and performance.
The variant preserves the core multimodal reasoning capabilities of its parent model, handling text, images, video, audio, and code with strong software engineering performance. With access to an extensive that quick-info value window, it excels at maintaining coherence across long interactions and complex tasks. This specialized version is designed for developers building automation pipelines, coding assistants, and complex workflow agents that depend on precise tool selection.
Transparent token rates
Rates are shown per one million tokens. Combined means one million input plus one million output tokens.
Vertex
Free Google Gemini 3.1 Pro Preview Custom Tools pricing calculator and token cost estimator. Count tokens, calculate API costs, and compare Gemini 3.1 Pro Preview Custom Tools pricing against GPT-5, Claude, Gemini & other models. Live token counter included.
Vertex
Magica's model profile confirms Gemini 3.1 Pro Preview Custom Tools as a Google-developed model with a 1M-token input context window and a 65,536-token maximum output. The page records a release date of February 25, 2026, and lists supported input types as text, audio, image, video, and file with text output using the The Magica page provides a detailed pricing breakdown for Gemini 3.1 Pro Preview Custom Tools: $2.00 per million input tokens and $12.00 per million output tokens at short context, plus long-context rates of $4.00 input and $18.00 output per million tokens for prompts at or above 200,000 tokens. Additional costs listed
Vertex
Models.dev lists Gemini 3.1 Pro Preview Custom Tools as an advanced Gemini model for complex reasoning, coding, and multimodal analysis, attributed to Google and belonging to the gemini-pro family. The registry entry confirms a context window of 1,048,576 tokens, an output limit of 65,536 tokens, a knowledge cutoff of According to the Models.dev catalog, Gemini 3.1 Pro Preview Custom Tools is served under the model ID gemini-3.1-pro-preview-customtools by Vertex (Google), alongside nine other providers including Databricks, Eden AI, Google, Kilo, Merge, NanoGPT, Neon, OpenRouter, and OrcaRouter. All listed providers share identical
This exact model name is also listed by 11 other providers.
