Gemma 4 31B IT
Gemma 4 31B IT is Google's open-weight dense model with 31B parameters, all active during inference. Built on the Gemini 3 architecture, it targets higher output quality than its MoE sibling, with support for function-calling, structured JSON output, native vision, and 140+ languages.
View API reference- Input and output price
- Prices from: Input $0.14, Output $0.40, Per 1M tokens
- 24h uptime
- Loading AI Gateway uptime
import { streamText } from 'ai'
const result = streamText({ model: 'google/gemma-4-31b-it', prompt: 'Why is the sky blue?'})Copy link to headingPlayground
Try out Gemma 4 31B IT by Google. Usage is billed to your team at API rates. Free users (those who haven't made a payment) get $5 of credits every 30 days.
Gemma 4 31B IT
Gemma 4 31B IT
Copy link to headingProviders
Route requests across multiple providers. Copy a provider slug to set your preference. Visit the docs for more info. Using a provider means you agree to their terms, listed under Legal.
| Provider |
|---|
| 262K | 131K | 0.8 s | 54 tps | $0.14/M | $0.40/M | — | +1 | 04/02/2026 | ||||||
| 262K | 131K | 0.6 s | 29 tps | $0.14/M | $0.40/M | — | 04/02/2026 | |||||||
| 256K | 131K | 7.9 s | 32 tps | $0.39/M | $0.97/M | — | +1 | 04/02/2026 | ||||||
| 131K | 40K | 0.2 s | 2479 tps | $0.99/M | $1.49/M | — | +1 | 04/02/2026 | ||||||
| 256K | 131K | 0.2 s | 122 tps | $0.14/M | $0.40/M | — | +1 | 04/02/2026 | ||||||
| 1M | 1M | $0.14/M | $0.40/M | Read$0.10/M | — | +1 | 04/02/2026 |
Copy link to headingUptime24 hours
Direct request success rate on AI Gateway and per-provider. Visit the docs for more info.
Copy link to headingThroughput24 hours
P50 throughput on live AI Gateway traffic, in tokens per second (TPS). Visit the docs for more info.
Copy link to headingLatency24 hours
P50 time to first token (TTFT) on live AI Gateway traffic, in milliseconds. View the docs for more info.
Getting started
Call Gemma 4 31B IT through AI Gateway with the AI SDK generateText and streamText functions, or through the OpenAI Chat Completions, OpenAI Responses, and Anthropic Messages APIs by changing the base URL. AI Gateway authenticates the request and routes it to an available provider.
Install the AI SDK (pnpm add ai dotenv), create an API key from the API Keys page, and set it as AI_GATEWAY_API_KEY in your environment. Full setup is covered in the text generation quickstart.
import { generateText } from 'ai';import 'dotenv/config';
async function main() { const result = await generateText({ model: 'google/gemma-4-31b-it', prompt: 'Why is the sky blue?', });
console.log(result.text);}
main().catch(console.error);Top-level parameters
The same Gemma 4 31B IT request in each API format AI Gateway supports.
import { generateText } from 'ai';import 'dotenv/config';
async function main() { const result = await generateText({ model: 'google/gemma-4-31b-it', system: 'You are a concise technical assistant.', prompt: 'Summarize the tradeoffs between static generation and SSR.', maxOutputTokens: 1024, });
console.log(result.text);}
main().catch(console.error);Standard parameters like prompt, messages, temperature, and tools work as documented in the AI SDK docs. These are the parameters with model-specific behavior.
| Parameter | Type | Required | Description |
|---|---|---|---|
model | string | Yes | Model ID in the form creator/model, e.g. google/gemma-4-31b-it. AI Gateway routes the request to an available provider. |
maxOutputTokens | number | No | Hard cap on generated tokens. Gemma 4 31B IT supports up to 1,048,576 output tokens. Reasoning tokens count toward this limit. |
reasoning | 'provider-default' | 'none' | 'minimal' | 'low' | 'medium' | 'high' | 'xhigh' | No | Provider-agnostic reasoning effort, available in AI SDK 7 or later. Maps to the provider’s native reasoning configuration; reasoning settings under providerOptions take precedence when both are set. See the Reasoning section below. |
providerOptions | Record<string, JSONValue> | No | AI Gateway routing options under gateway, plus any provider-native options under the provider’s own namespace — see the table below. |
Input limits
| Input | Formats | Sources | Max count | Max size | Limits |
|---|---|---|---|---|---|
| Text | — | — | — | — | Prompt and response share the 1M-token context window |
| Image | — | URL, base64, Uint8Array | — | — | Sent as image parts in messages; counts as input tokens |
| — | URL, base64, Uint8Array | — | — | Sent as file parts in messages; counts as input tokens |
Provider options
Set AI Gateway routing options under providerOptions.gateway. For provider-specific options, pass them under the provider’s namespace as documented by the AI SDK.
Learn more in the AI SDK google provider docs.
import { generateText } from 'ai';import 'dotenv/config';
async function main() { const result = await generateText({ model: 'google/gemma-4-31b-it', prompt: 'Why is the sky blue?', providerOptions: { gateway: { only: ['novita', 'parasail'], }, }, });
console.log(result.text);}
main().catch(console.error);These AI Gateway routing options apply to every model. Provider-specific options pass through under the provider’s own namespace (for example providerOptions.anthropic) exactly as documented by the AI SDK.
| Parameter | Type | Required | Description |
|---|---|---|---|
providerOptions.gateway.only | string[] | No | Restrict routing to these provider slugs. Requests fail over only within the listed providers. |
providerOptions.gateway.order | string[] | No | Preferred provider order. Listed providers are tried first; unlisted providers remain available as fallbacks. |
providerOptions.gateway.sort | 'cost' | 'ttft' | 'tps' | No | Rank candidate providers by price, time to first token, or tokens per second instead of the default routing order. |
providerOptions.gateway.zeroDataRetention | boolean | No | Route only to providers with a zero-data-retention policy for this model. |
Routing across providers
AI Gateway serves the same model through multiple providers and fails over automatically. order expresses a preference while keeping every provider eligible; only is a hard allowlist — if none of the listed providers are available the request fails instead of falling back.
Options under a provider's own namespace (for example providerOptions.anthropic) are forwarded to that provider with the request. Providers ignore option namespaces that don't apply to them, so it is safe to set provider options alongside gateway routing options.
Reasoning
AI Gateway bridges reasoning across every API format. The AI SDK exposes a provider-agnostic top-level reasoning level (none, minimal, low, medium, high, or xhigh); the Chat Completions and Responses formats take the same effort under reasoning.effort; and the Anthropic Messages format uses a native thinking token budget. Whichever you send, the gateway maps it to the target model’s native configuration, converting between effort levels and token budgets as needed. Reasoning-related settings under providerOptions take full precedence over the top-level reasoning value and are never merged. Reasoning tokens typically count toward your output-token usage, though how they’re reported and billed varies by provider.
Learn more in the AI Gateway reasoning guide.
import { generateText } from 'ai';import 'dotenv/config';
async function main() { const result = await generateText({ model: 'google/gemma-4-31b-it', prompt: 'Explain the Monty Hall problem step by step.', reasoning: 'high', });
console.log(result.text);}
main().catch(console.error);Image input
Send images alongside text as message parts. Images count as input tokens.
import { generateText } from 'ai';import 'dotenv/config';
async function main() { const result = await generateText({ model: 'google/gemma-4-31b-it', messages: [ { role: 'user', content: [ { type: 'text', text: 'Describe this image.' }, { type: 'image', image: 'https://example.com/photo.jpg' }, ], }, ], });
console.log(result.text);}
main().catch(console.error);PDF input
Attach PDFs as file parts. Their contents count as input tokens.
import { generateText } from 'ai';import 'dotenv/config';
async function main() { const result = await generateText({ model: 'google/gemma-4-31b-it', messages: [ { role: 'user', content: [ { type: 'text', text: 'Summarize this document.' }, { type: 'file', mediaType: 'application/pdf', data: 'https://example.com/document.pdf', }, ], }, ], });
console.log(result.text);}
main().catch(console.error);Tool calling
Expose tools the model can call. Define each tool’s inputs with a Zod schema.
import { generateText, tool } from 'ai';import { z } from 'zod';import 'dotenv/config';
async function main() { const result = await generateText({ model: 'google/gemma-4-31b-it', prompt: 'What is the weather in San Francisco?', tools: { getWeather: tool({ description: 'Get the current weather for a location', inputSchema: z.object({ location: z.string() }), execute: async ({ location }) => ({ location, temperatureC: 18 }), }), }, });
console.log(result.text);}
main().catch(console.error);Copy link to headingMore models by Google
| Model |
|---|
| 1M | 1.9 s | 417 tps | $0.75/M | $3.75/M | Read$0.08/M | $14/K+1 more | +3 |  | 09/02/2026 | |||||
| 1M | 0.9 s | 279 tps | $0.75/M | $3.75/M | Read$0.08/M | $14/K+1 more | +3 | | 08/13/2026 | |||||
| 1M | 0.5 s | 307 tps | $0.30/M | $2.50/M | Read$0.03/M | $14/K+1 more | +3 | | 07/21/2026 | |||||
| 1M | 0.6 s | 255 tps | $0.25/M | $1.50/M | Read$0.03/M | $14/K+1 more | +3 | | 05/07/2026 | |||||
| 1M | 0.6 s | 199 tps | $0.50/M+1 more | $3/M+1 more | Read$0.05/M | $14/K+1 more | +3 | | 12/17/2025 | |||||
| 1M | 0.3 s | 271 tps | $0.10/M | $0.40/M | Read$0.01/M | $35/K+1 more | +3 | | 06/17/2025 |
Copy link to headingAbout Gemma 4 31B IT
Gemma 4 31B IT is the dense counterpart in Google's Gemma 4 family, released on April 2, 2026 alongside the mixture-of-experts Gemma 4 26B. While both share the Gemini 3 architecture, this model activates all 31B parameters during every forward pass.
The dense design means every parameter contributes to every prediction. This produces higher output quality on complex reasoning, generation, and analysis tasks compared to the MoE variant, where a routing mechanism selects a subset of parameters. The tradeoff is higher compute per token, which translates to increased latency and cost per request.
Gemma 4 31B IT accepts text and image inputs within a context window of 1.0M tokens, supports over 140 languages, and handles function-calling, agentic workflows, structured JSON output, and system instructions. The instruction-tuning (indicated by the it suffix) prepares the model for conversational and task-oriented use out of the box.
Running Gemma 4 31B IT through AI Gateway provides unified billing, observability, automatic retries, and provider failover across a single API surface.
Copy link to headingWhat To Consider When Choosing a Provider
- Configuration: As a dense model with all parameters active, Gemma 4 31B IT uses more compute per token than the MoE Gemma 4 26B variant. Factor in the higher per-request cost and latency when evaluating provider variants for production traffic.
- Zero Data Retention: Zero Data Retention is available for this model. It is offered on a per-provider and model basis. See the documentation for details.
- Authentication: AI Gateway authenticates requests using an API key or OIDC token. You do not need to manage provider credentials directly.
Copy link to headingWhen to Use Gemma 4 31B IT
Best for
- Quality-critical generation tasks: You need the strongest output in the Gemma 4 family and can accept higher latency
- Complex reasoning and analysis: Multi-step planning, code generation, and detailed document analysis
- Multilingual applications: Serving users across 140+ languages with a single model
- Vision-language tasks: Image understanding, visual Q&A, and document parsing within a context window of 1.0M tokens
Consider alternatives when
- Latency and throughput primary: Your primary constraints favor the MoE Gemma 4 26B, which activates fewer parameters and responds faster
- Native image or audio generation: You need media output, which Gemma 4 31B IT does not support
- High-volume low-complexity inference: A smaller or lighter model is more cost-effective
- Proprietary-grade benchmark performance: Gemini 3 Pro may be a better fit for the most demanding benchmarks
Copy link to headingConclusion
Gemma 4 31B IT is the quality-focused option in the Gemma 4 family. With all 31B parameters active during inference, it delivers stronger output on complex tasks. For teams that want open-weight flexibility with the highest reasoning quality the Gemma 4 generation offers, it is the right starting point on AI Gateway.
Copy link to headingFrequently Asked Questions
What makes Gemma 4 31B IT different from the MoE Gemma 4 26B?
Gemma 4 31B IT is a dense model, meaning all 31B parameters are active during every forward pass. The MoE Gemma 4 26B activates roughly 4B of its 26B total parameters per pass. Gemma 4 31B IT targets higher output quality; the 26B variant targets lower latency and cost.
What input modalities does Gemma 4 31B IT support?
Gemma 4 31B IT accepts text and image inputs within a context window of 1.0M tokens. It does not generate images or audio.
How does Gemma 4 31B IT relate to Google's Gemini models?
Gemma 4 31B IT is built on the same architecture as Gemini 3 but with open weights. It shares capabilities like function-calling, structured output, and system instructions. Gemini models remain proprietary; Gemma 4 31B IT lets you inspect or adapt the weights.
What languages does Gemma 4 31B IT support?
Over 140 languages. The instruction-tuning covers multilingual conversational and task-oriented use cases.
How do I use Gemma 4 31B IT on AI Gateway?
Set the model to
google/gemma-4-31b-itin the AI SDK. AI Gateway handles provider routing, retries, and failover automatically.Does Gemma 4 31B IT support function-calling?
Yes. It supports function-calling for agentic workflows, structured JSON output, and system instructions natively, inherited from the Gemini 3 architecture.