Current CyVerse Models¶
Important
Last Updated: July 24, 2025
This page is provides details about models provided directly through CyVerse. Information about models from other national inference infrastructure can be referenced through the respective documentation or informational web sites.
Model Name | Aliases | Token Context Length | Supports Vision | Supports Reasoning | Supports Tools |
---|---|---|---|---|---|
gemma-3-12b-it | 32,768 | ||||
Llama-3.3-70B-Instruct-quantized | 128,000 | ||||
Llama-3.2-11B-Vision-Instruct | 32,768 | ||||
phi-4-multimodal-instruct | phi-4 | 128,000 | |||
qwen3-coder-32b | 131,072 |
Note
To use a model, you can use the model name or alias (if provided.)
Deprecated Models¶
This section lists models that have been deprecated through "cyverse".
- Qwen2.5-Coder-32B-Instruct, removed on July 20, 2025