Google releases Gemma 4 31B free model with 256K context and multimodal support
Google DeepMind has released Gemma 4 31B Instruct, a free 30.7-billion parameter model with a 256K token context window, multimodal text and image input capabilities, and native function calling. The model supports configurable reasoning mode and 140+ languages, with strong performance on coding and document understanding tasks under Apache 2.0 license.
Gemma 4 31B Instruct — Quick Specs
Google Releases Gemma 4 31B Free Model with Multimodal Support
Google DeepMind has released Gemma 4 31B Instruct as a free, open-source model available via OpenRouter as of April 2, 2026. The 30.7-billion parameter dense model introduces multimodal capabilities, accepting both text and image inputs while outputting text.
Key Specifications
Context Window & Performance
- 256,144 token context window
- Configurable thinking/reasoning mode for step-by-step problem solving
- Native function calling support
- Multilingual capability across 140+ languages
Availability & Pricing
- Free to use with $0 per million input tokens and $0 per million output tokens
- Available under Apache 2.0 license
- Accessible via OpenRouter's API with model weights available for download
- OpenRouter routes requests across multiple providers with automatic fallback for uptime
Capabilities
Google DeepMind claims Gemma 4 31B demonstrates strong performance on coding tasks, reasoning-heavy problems, and document understanding. The configurable reasoning mode allows users to enable step-by-step thinking processes, with reasoning details accessible in API responses for transparency into the model's internal logic.
The model supports function calling natively, enabling integration with external tools and APIs. Its 140+ language support indicates broader multilingual capability compared to earlier Gemma versions.
Technical Implementation
OpenRouter's infrastructure routes requests to optimal providers based on prompt size and parameters, with fallback systems to maximize availability. Users can enable reasoning through API parameters and preserve reasoning details across conversation turns to maintain context continuity.
Model weights are available for local deployment, giving developers options for both API-based and self-hosted usage. The Apache 2.0 license permits commercial and research applications without restriction.
What This Means
Gemma 4 31B's free release with 256K context and multimodal support directly challenges proprietary models in the mid-range segment. The zero-cost pricing and open license make it viable for cost-sensitive production deployments. The reasoning mode addition and 140+ language support suggest Google is competing on capability breadth rather than just scale. For organizations currently paying for Claude or GPT-4 access on routine tasks, this release provides a credible alternative worth evaluation—particularly for coding, document analysis, and multilingual workloads.
Related Articles
Nex AGI releases Nex-N2-Mini: open-source agentic MoE model with 262K context window
Nex AGI has released Nex-N2-Mini, an open-source agentic mixture-of-experts model with a 262K-token context window. The model accepts text and image inputs and is priced at $0.025 per 1M input tokens and $0.10 per 1M output tokens.
Mistral releases Leanstral 1.5: 119B parameter open-source model for Lean 4 proof assistance
Mistral AI has released Leanstral 1.5, an open-source 119B parameter mixture-of-experts model designed specifically for Lean 4 proof assistance. The model features 128 experts with 4 active per token (6.5B activated parameters), a 256k token context window, and multimodal input capabilities.
Portugal releases Amália, open-source 9B parameter AI model trained on European Portuguese
Portugal has released Amália, its first national AI model trained specifically for European Portuguese. Built on EuroLLM-9B with 9 billion parameters, the model is fully open-source with weights, datasets, and code published under an open license. The government has committed €5.5m in initial funding through 2027.
DeepReinforce Releases Ornith-1.0, Open-Source Agentic Coding Model in 9B to 397B Sizes
DeepReinforce has released Ornith-1.0, an MIT-licensed model designed for agentic coding tasks with variants ranging from 9B to 397B parameters. Built on top of Apache 2.0-licensed Gemma 4 and Qwen 3.5 base models, the company claims it achieves state-of-the-art performance among open-source models of comparable size on coding benchmarks.
Comments
Loading...