fix: add support for min_tokens

Signed-off-by: Aaron Pham <29749331+aarnphm@users.noreply.github.com>

openllm-python/src/_openllm_tiny/_llm.py

@@ -171,6 +171,7 @@ class LLM:
 
     top_p = 1.0 if config['temperature'] <= 1e-5 else config['top_p']
     config = config.model_copy(update=dict(stop=list(stop), stop_token_ids=stop_token_ids, top_p=top_p))
+
     sampling_params = SamplingParams(**{
       k: getattr(config, k, None) for k in set(inspect.signature(SamplingParams).parameters.keys())
     })

openllm-python/src/_openllm_tiny/_service.py

@@ -92,7 +92,7 @@ class LLMService:
       stop_token_ids=stop_token_ids,
       request_id=request_id,
     ):
-      yield f'data: {generated.model_dump_json()}\n\n'
+      yield f'data: {core.GenerationOutput.from_vllm(generated).model_dump_json()}\n\n'
     yield 'data: [DONE]\n\n'
 
   @core.utils.api(route='/v1/metadata')