Parse the model string to extract base model and adapter IDs
(model: str)
| 2142 | return InferenceClient() |
| 2143 | |
| 2144 | def parse_model_string(model: str) -> ModelConfig: |
| 2145 | """Parse the model string to extract base model and adapter IDs""" |
| 2146 | parts = model.split('+') |
| 2147 | base_model_id = parts[0] |
| 2148 | adapter_ids = parts[1:] if len(parts) > 1 else None |
| 2149 | |
| 2150 | return ModelConfig( |
| 2151 | base_model_id=base_model_id, |
| 2152 | adapter_ids=adapter_ids, |
| 2153 | use_memory_efficient_attention=False, |
| 2154 | quantization_bits=0, |
| 2155 | enable_prompt_caching=False, |
| 2156 | dynamic_temperature=False, |
| 2157 | ) |
| 2158 | |
| 2159 | def get_effort_profile(reasoning_effort: str, max_tokens: int = 4096) -> dict: |
| 2160 | """Get reasoning effort profile based on specified level and max tokens. |
no test coverage detected