Gemma 3.2
Gemma 3.2 is a 27-billion-parameter text generation model from Google with a 128,000-token context window.
Google
DeepInfraGoogle's open text model with long context
Gemma 3.2 is an instruction-tuned text generation model developed by Google, released in March 2025 under the full identifier google/gemma-3-27b-it. It belongs to the Gemma 3 family of open models and is served here via DeepInfra with a 128,000-token context window and a maximum response size of 8,000 tokens. The model is tagged for reasoning and multi-modal capabilities, reflecting its design to handle a range of language understanding and generation tasks.
With 27 billion parameters, Gemma 3.2 is suited for tasks that benefit from a large context window, such as document summarization, multi-turn conversation, and complex reasoning over long inputs. It follows an instruction-tuned format, meaning it is optimized to respond to user prompts and directives rather than function as a raw completion model. Developers looking for an open-weight model from Google that can handle extended context and reasoning tasks will find this a practical option within MindStudio.
What Gemma 3.2 supports
Long Context Window
Processes up to 128,000 tokens in a single request, enabling analysis of lengthy documents, codebases, or extended conversations without truncation.
Instruction Following
Trained in an instruction-tuned format (the -it suffix), the model is optimized to respond accurately to explicit user prompts and multi-turn directives.
Reasoning
Tagged for reasoning tasks, the model can work through multi-step problems, logical inference, and structured analysis within a single context.
Multi-Modal Awareness
Tagged as multi-modal, indicating the model architecture supports inputs beyond plain text, consistent with the broader Gemma 3 family design.
Text Generation
Generates coherent, contextually grounded text across formats including summaries, explanations, and conversational replies, with responses up to 8,000 tokens.
Ready to build with Gemma 3.2?
Get Started FreeBenchmark scores
Scores represent accuracy — the percentage of questions answered correctly on each test.
| Benchmark | What it tests | Score |
|---|---|---|
| MMLU-Pro | Expert knowledge across 14 academic disciplines | 66.9% |
| GPQA Diamond | PhD-level science questions (biology, physics, chemistry) | 42.8% |
| MATH-500 | Undergraduate and competition-level math problems | 88.3% |
| AIME 2024 | American math olympiad problems | 25.3% |
| LiveCodeBench | Real-world coding tasks from recent competitions | 13.7% |
| HLE | Questions that challenge frontier models across many domains | 4.7% |
| SciCode | Scientific research coding and numerical methods | 21.2% |
Common questions about Gemma 3.2
What is the context window for Gemma 3.2?
Gemma 3.2 supports a context window of 128,000 tokens, allowing it to process long documents or extended conversations in a single request.
What is the maximum response length?
The model has a maximum response size of 8,000 tokens per output.
Who developed Gemma 3.2 and when was it released?
Gemma 3.2 was developed by Google and released in March 2025. It is served on MindStudio via the DeepInfra provider.
Is pricing available for this model on MindStudio?
No published pricing is listed in the current metadata for this model. Check MindStudio's pricing page or your account dashboard for current usage costs.
Is Gemma 3.2 an open-weight model?
Yes. The Gemma family from Google is released as open-weight models. The specific variant here is google/gemma-3-27b-it, the 27-billion-parameter instruction-tuned version.
What is the knowledge cutoff for Gemma 3.2?
The metadata does not specify a knowledge cutoff date. Based on its March 2025 release, training data likely extends into late 2024, but the exact cutoff has not been confirmed in the available information.
Parameters & options
Explore similar models
GoogleGemma 4 31B
Gemma 4 31B is a multimodal text generation model from Google with a 262,144-token context window.
GoogleGemini 3.1 Pro
Gemini 3.1 Pro is a multimodal text generation model from Google with a 1,048,576-token context window.
GoogleGemini 2.5 Pro
Gemini 2.5 Pro is Google's flagship multimodal model with a 1,000,000-token context window and built-in reasoning capabilities.
QwenQwen3.8 27B
Step 3.7 Flash
AnthropicClaude 5 Sonnet
Claude 5 Sonnet is Anthropic's flagship text and vision model with a 1,000,000-token context window and extended reasoning.
Start building with Gemma 3.2
No API keys required. Create AI-powered workflows with Gemma 3.2 in minutes — free.