Model Type | Generative, Multilingual, Conversational AI |
|
Use Cases |
Areas: | Research, Commercial applications |
|
Applications: | Reasoning, Summarization, Question answering, Multilingual generation, Conversational AI |
|
|
Additional Notes | Command-R has tool use and grounded generation capabilities. Suitable for interaction with code. |
|
Supported Languages | en (Supported), fr (Supported), es (Supported), it (Supported), de (Supported), pt (Supported), ja (Supported), ko (Supported), zh (Supported), ar (Supported), ru (Pre-trained), pl (Pre-trained), tr (Pre-trained), vi (Pre-trained), nl (Pre-trained), cs (Pre-trained), id (Pre-trained), uk (Pre-trained), ro (Pre-trained), el (Pre-trained), hi (Pre-trained), he (Pre-trained), fa (Pre-trained) |
|
Training Details |
Data Sources: | English, French, Spanish, Italian, German, Brazilian Portuguese, Japanese, Korean, Simplified Chinese, and Arabic, Russian, Polish, Turkish, Vietnamese, Dutch, Czech, Indonesian, Ukrainian, Romanian, Greek, Hindi, Hebrew, Persian |
|
Methodology: | Supervised fine-tuning and preference training |
|
Context Length: | |
Model Architecture: | Auto-regressive transformer |
|
|
Input Output |
Input Format: | |
Accepted Modalities: | |
Output Format: | |
Performance Tips: | Use the specified template for best performance. Encourage experimentation with the prompt template. |
|
|