Gemini 4 Neon delivers high-performance coding, multi-step reasoning, and multimodal understanding at low latency for agentic enterprise workflows. Gemini 4 Neon is a Google Frontier Model subject to Section 34 of the Service Specific Terms (for details, see Abuse monitoring).
Neon differs from earlier Gemini models in the following ways:
Unified input and output token limit and long decodes: Neon shares a unified 1,048,576-token limit across input and output tokens on the
GenerateContentAPI, the Interactions API, and the OpenAI compatibility Chat Completions API. Output tokens include thinking, function calls, and the final response, and a single foreground segment generates up to 65,536 output tokens. On the Interactions API andGenerateContent(v1beta1), long decodes extend output up to 1,048,576 tokens minus your input tokens. The Google Gen AI SDK and Agent Studio followcontinuation_tokenautomatically across segments (v1beta1by default). You can also runbackground=trueon the Interactions API or passcontinuation_tokenin a custom loop. For details, see the developer's guide to Gemini 4 Neon.Thinking levels: Neon supports thinking levels
LOW,MEDIUM,HIGH(default), andXHIGH. Settingthinking_budgetorthinking_level="MINIMAL"returns an HTTP400validation error.Sampling parameters: Custom values for sampling parameters (
temperature,top_p, andtop_k) are ignored.Context caching: Neon supports implicit caching only (4,096-token minimum), with durable caching enabled by default. Explicit context caching isn't supported. Starting on January 12, 2027, Gemini 4 Hydrogen and later models support cache control features with TTL configuration. You choose what to cache and how long cached content stays valid through TTL configuration. Cache writes are billed at 1.25x the standard input token price by default, and longer retention costs more. For details, see Gemini 4 pricing.
Office and OpenDocument file inputs: In addition to PDF (
application/pdf) and plain text (text/plain), Neon accepts Microsoft Office (.docx,.dotx,.doc,.pptx,.ppt,.xlsx,.xls) and OpenDocument (.odt,.odp,.ods) files inline or from Cloud Storage on both theGenerateContentAPI and the Interactions API, including inside multimodalFunctionResponseparts.
Try in Agent Studio Deploy example app Developer guide Pricing
| Model ID | gemini-4-neon |
|
|---|---|---|
| Modalities |
|
|
| Token limits | Context window | 1,048,576 |
| Maximum output tokens | 1,048,576 | |
| Capabilities |
|
|
| Tools |
|
|
| APIs |
|
|
| Consumption options |
|
|
| Input size limit |
|
|
| Interactions API URI limit |
Up to 100 distinct Cloud Storage (gs://) URIs per request across all modalities on the Interactions API, including URIs carried forward from earlier turns using previous_interaction_id. generateContent supports gs:// URIs up to the per-modality model item limits (up to 6,500 combined multimodal files when including plain-text files).
|
|
| Technical specifications | Image |
|
| Text |
|
|
| Video |
|
|
| Audio |
|
|
| Parameter defaults |
|
|
| Supported regions |
|
|
|
||
|
||
|
||
|
||
| Knowledge cutoff date | March 2026 | |
| Versions |
|
|
| Security controls | Online prediction |
|
| Batch inference |
|
|
| Context caching |
|
|
| See Security controls for more information. | ||