Gemma 3 27B (free) - Complete Model Details
Overview
Model Name: Google Gemma 3 27B (free)
Creator: Google
Model ID: google/gemma-3-27b-it:free
Permaslug: google/gemma-3-27b-it
Released: March 12, 2025
Description: This model introduces multimodal capabilities with vision-language input and text outputs. It handles context windows up to 128k tokens and understands over 140 languages with enhanced math and reasoning capabilities including structured outputs and function calling.
Pricing
| Component | Cost |
|---|---|
| Input Tokens | $0/M |
| Output Tokens | $0/M |
| Images | $0 |
| Requests | $0 |
| Web Search | $0 |
Pricing Model: Completely free
Performance
Usage Statistics (Last 30 Days)
- Daily Requests: 67,000-99,000 requests
- Completion Tokens: 34-64 million tokens daily
- Prompt Tokens: 267-505 million tokens daily
- Image Processing: 1,400-60,000 images processed daily
Technical Specifications
- Context Length: 131,072 tokens (128k)
- Input Modalities: Text, Image
- Output Modalities: Text only
- Instruct Type: Gemma
- Model Group: Gemini
- Reasoning Support: Not supported
- Default Stop Sequences:
<start_of_turn>,<end_of_turn>,<eos>
Related Models
- Predecessor: Gemma 2 27B (google/gemma-2-27b-it)
Providers
Primary Provider: ModelRun
| Attribute | Value |
|---|---|
| Context Window | 128,000 tokens |
| Adapter | OpenAIAdapter |
| Base URL | https://api.langmart.ai/v1 |
| Headquarters | US |
| Datacenters | US |
| Chat Completions | Supported |
| Text Completions | Supported |
| Abortable Requests | Yes |
| Multipart Support | Yes |
Parameters
Supported request parameters:
- max_tokens
- temperature
- presence_penalty
- repetition_penalty
- frequency_penalty
Data Policy
| Policy | Value |
|---|---|
| Training Data Usage | Not allowed |
| Prompt Retention | No |
| Publishable Results | No |
| Terms of Service | https://www.runmodelrun.com/TOS.html |
| Privacy Policy | https://www.runmodelrun.com/privacy-policy.html |
Available Endpoints
- Chat Completions
- Model Comparison