DeepSeek reasoning model for multi-step analysis, math, coding, and tools
Details
| Field | Value |
|---|---|
| id | deepseek-r1-distill-llama-70b |
| family | deepseek-thinking |
| modality | text->text |
| context_length | 32678 |
| max_output_tokens | 8192 |
| release_date | 2025-01-30 |
| is_open_weights | yes |
| supports_tool_call | yes |
| supports_reasoning | yes |
| supports_structured_output | yes |
| supports_attachment | no |
| input_modalities | text |
| output_modalities | text |