Merge pull request #224 from restackio/update-llm-model

leawn · web-flow · commit d886f7c521a9 · 2025-04-26T13:19:53.000+02:00
Update llm model to latest available
diff --git a/agent_apis/src/functions/llm.py b/agent_apis/src/functions/llm.py
@@ -41,7 +41,7 @@ async def llm(function_input: FunctionInputParams) -> str:
         messages.append({"role": "user", "content": function_input.user_content})
 
         response = client.chat.completions.create(
-            model=function_input.model or "gpt-4o-mini", messages=messages
+            model=function_input.model or "gpt-4.1-mini", messages=messages
         )
         log.info("llm function completed", response=response)
         return response.choices[0].message.content
diff --git a/agent_apis/src/workflows/multistep.py b/agent_apis/src/workflows/multistep.py
@@ -35,7 +35,7 @@ async def run(self, workflow_input: WorkflowInputParams) -> dict:
                     function_input=FunctionInputParams(
                         system_content=f"You are a personal assitant and have access to weather data {weather_data}. Always greet person with relevant info from weather data",
                         user_content=user_content,
-                        model="gpt-4o-mini",
+                        model="gpt-4.1-mini",
                     ),
                     start_to_close_timeout=timedelta(seconds=120),
                 )
diff --git a/agent_chat/src/functions/llm_chat.py b/agent_chat/src/functions/llm_chat.py
@@ -44,7 +44,7 @@ async def llm_chat(agent_input: LlmChatInput) -> dict[str, str]:
             )
 
         assistant_raw_response = client.chat.completions.create(
-            model=agent_input.model or "gpt-4o-mini",
+            model=agent_input.model or "gpt-4.1-mini",
             messages=agent_input.messages,
         )
     except Exception as e:
diff --git a/agent_rag/README.md b/agent_rag/README.md
@@ -51,7 +51,7 @@ python -c "from src.services import watch_services; watch_services()"
 
 Duplicate the `env.example` file and rename it to `.env`.
 
-Obtain a Restack API Key to interact with the 'gpt-4o-mini' model at no cost from [console.restack.io](https://console.restack.io)
+Obtain a Restack API Key to interact with the 'gpt-4.1-mini' model at no cost from [console.restack.io](https://console.restack.io)
 
 ## Run agents
 
diff --git a/agent_rag/src/functions/llm_chat.py b/agent_rag/src/functions/llm_chat.py
@@ -45,7 +45,7 @@ async def llm_chat(function_input: LlmChatInput) -> ChatCompletion:
             )
 
         response = client.chat.completions.create(
-            model=function_input.model or "gpt-4o-mini",
+            model=function_input.model or "gpt-4.1-mini",
             messages=function_input.messages,
         )
     except Exception as e:
diff --git a/agent_stream/src/functions/llm_chat.py b/agent_stream/src/functions/llm_chat.py
@@ -40,7 +40,7 @@ async def llm_chat(function_input: LlmChatInput) -> str:
         messages_dicts = [message.model_dump() for message in function_input.messages]
         # Get the streamed response from OpenAI API
         response: Stream[ChatCompletionChunk] = client.chat.completions.create(
-            model=function_input.model or "gpt-4o-mini",
+            model=function_input.model or "gpt-4.1-mini",
             messages=messages_dicts,
             stream=True,
         )
diff --git a/agent_telephony/twilio_livekit/readme.md b/agent_telephony/twilio_livekit/readme.md
@@ -31,7 +31,7 @@ docker run -d --pull always --name restack -p 5233:5233 -p 6233:6233 -p 7233:723
 
 In all subfolders, duplicate the `env.example` file and rename it to `.env`.
 
-Obtain a Restack API Key to interact with the 'gpt-4o-mini' model at no cost from [Restack Cloud](https://console.restack.io/starter)
+Obtain a Restack API Key to interact with the 'gpt-4.1-mini' model at no cost from [Restack Cloud](https://console.restack.io/starter)
 
 
 ## Start Restack Agent with Twilio
@@ -102,7 +102,7 @@ python src/worker.py dev
 
 Duplicate the `env.example` file and rename it to `.env`.
 
-Obtain a Restack API Key to interact with the 'gpt-4o-mini' model at no cost from [Restack Cloud](https://console.restack.io/starter)
+Obtain a Restack API Key to interact with the 'gpt-4.1-mini' model at no cost from [Restack Cloud](https://console.restack.io/starter)
 
 ## Create a new Agent
 
diff --git a/agent_telephony/vapi/agent_vapi/readme.md b/agent_telephony/vapi/agent_vapi/readme.md
@@ -28,7 +28,7 @@ docker run -d --pull always --name restack -p 5233:5233 -p 6233:6233 -p 7233:723
 
 In all subfolders, duplicate the `env.example` file and rename it to `.env`.
 
-Obtain a Restack API Key to interact with the 'gpt-4o-mini' model at no cost from [Restack Cloud](https://console.restack.io/starter)
+Obtain a Restack API Key to interact with the 'gpt-4.1-mini' model at no cost from [Restack Cloud](https://console.restack.io/starter)
 
 
 ## Start Restack Agent with Twilio
@@ -99,7 +99,7 @@ python src/pipeline.py dev
 
 Duplicate the `env.example` file and rename it to `.env`.
 
-Obtain a Restack API Key to interact with the 'gpt-4o-mini' model at no cost from [Restack Cloud](https://console.restack.io/starter)
+Obtain a Restack API Key to interact with the 'gpt-4.1-mini' model at no cost from [Restack Cloud](https://console.restack.io/starter)
 
 ## Create a new Agent
 
diff --git a/agent_telephony/vapi/agent_vapi/src/functions/llm_chat.py b/agent_telephony/vapi/agent_vapi/src/functions/llm_chat.py
@@ -40,7 +40,7 @@ async def llm_chat(function_input: LlmChatInput) -> str:
         messages_dicts = [message.model_dump() for message in function_input.messages]
         # Get the streamed response from OpenAI API
         response: Stream[ChatCompletionChunk] = client.chat.completions.create(
-            model=function_input.model or "gpt-4o-mini",
+            model=function_input.model or "gpt-4.1-mini",
             messages=messages_dicts,
             stream=True,
         )
diff --git a/agent_todo/src/functions/llm_chat.py b/agent_todo/src/functions/llm_chat.py
@@ -55,7 +55,7 @@ async def llm_chat(function_input: LlmChatInput) -> ChatCompletion:
             )
 
         response = client.chat.completions.create(
-            model=function_input.model or "gpt-4o-mini",
+            model=function_input.model or "gpt-4.1-mini",
             messages=function_input.messages,
             tools=function_input.tools,
         )
diff --git a/agent_tool/README.md b/agent_tool/README.md
@@ -51,7 +51,7 @@ python -c "from src.services import watch_services; watch_services()"
 
 Duplicate the `env.example` file and rename it to `.env`.
 
-Obtain a Restack API Key to interact with the 'gpt-4o-mini' model at no cost from [Restack Cloud](https://console.restack.io/starter)
+Obtain a Restack API Key to interact with the 'gpt-4.1-mini' model at no cost from [Restack Cloud](https://console.restack.io/starter)
 
 ## Run agents
 
diff --git a/agent_tool/src/functions/llm_chat.py b/agent_tool/src/functions/llm_chat.py
@@ -55,7 +55,7 @@ async def llm_chat(function_input: LlmChatInput) -> ChatCompletion:
             )
 
         result = client.chat.completions.create(
-            model=function_input.model or "gpt-4o-mini",
+            model=function_input.model or "gpt-4.1-mini",
             messages=function_input.messages,
             tools=function_input.tools,
         )
diff --git a/agent_video/README.md b/agent_video/README.md
@@ -55,7 +55,7 @@ python -c "from src.services import watch_services; watch_services()"
 
 Duplicate the `env.example` file and rename it to `.env`.
 
-Obtain a Restack API Key to interact with the 'gpt-4o-mini' model at no cost from [Restack Cloud](https://console.restack.io/starter)
+Obtain a Restack API Key to interact with the 'gpt-4.1-mini' model at no cost from [Restack Cloud](https://console.restack.io/starter)
 
 ## Create Room and run Agent in parallel
 
diff --git a/agent_video/src/functions/llm_chat.py b/agent_video/src/functions/llm_chat.py
@@ -40,7 +40,7 @@ async def llm_chat(function_input: LlmChatInput) -> str:
         messages_dicts = [message.model_dump() for message in function_input.messages]
         # Get the streamed response from OpenAI API
         response: Stream[ChatCompletionChunk] = client.chat.completions.create(
-            model=function_input.model or "gpt-4o-mini",
+            model=function_input.model or "gpt-4.1-mini",
             messages=messages_dicts,
             stream=True,
         )
diff --git a/agent_voice/livekit/README.md b/agent_voice/livekit/README.md
@@ -34,7 +34,7 @@ In all subfolders, duplicate the `env.example` file and rename it to `.env`.
   - Sign up at [ElevenLabs](https://elevenlabs.io/docs/overview)
   - Add `ELEVENLABS_API_KEY` to .env file
 
-Obtain a Restack API Key to interact with the 'gpt-4o-mini' model at no cost from [Restack Cloud](https://console.restack.io/starter)
+Obtain a Restack API Key to interact with the 'gpt-4.1-mini' model at no cost from [Restack Cloud](https://console.restack.io/starter)
 
 ## Start Restack
 
diff --git a/agent_voice/livekit/agent/src/functions/llm_chat.py b/agent_voice/livekit/agent/src/functions/llm_chat.py
@@ -40,7 +40,7 @@ async def llm_chat(function_input: LlmChatInput) -> str:
         messages_dicts = [message.model_dump() for message in function_input.messages]
         # Get the streamed response from OpenAI API
         response: Stream[ChatCompletionChunk] = client.chat.completions.create(
-            model=function_input.model or "gpt-4o-mini",
+            model=function_input.model or "gpt-4.1-mini",
             messages=messages_dicts,
             stream=True,
         )
diff --git a/agent_voice/livekit/livekit_pipeline/src/pipeline.py b/agent_voice/livekit/livekit_pipeline/src/pipeline.py
@@ -95,7 +95,7 @@ async def entrypoint(ctx: JobContext) -> None:
         vad=ctx.proc.userdata["vad"],
         stt=deepgram.STT(),
         llm=openai.LLM(
-            # model="gpt-4o-mini",
+            # model="gpt-4.1-mini",
             # api_key=os.environ.get("OPENAI_API_KEY"),
             api_key=f"{agent_id}-livekit",
             base_url=agent_url,
diff --git a/agent_voice/pipecat/pipecat_pipeline/src/pipeline.py b/agent_voice/pipecat/pipecat_pipeline/src/pipeline.py
@@ -51,7 +51,7 @@ async def main() -> None:
         live_options=LiveOptions(vad_events=True, utterance_end_ms="1000"),
     )
 
-    llm = OpenAILLMService(api_key=os.getenv("OPENAI_API_KEY"), model="gpt-4o-mini")
+    llm = OpenAILLMService(api_key=os.getenv("OPENAI_API_KEY"), model="gpt-4.1-mini")
 
     tts = ElevenLabsTTSService(
         api_key=os.getenv("ELEVENLABS_API_KEY", ""),
diff --git a/audio_transcript/pyproject.toml b/audio_transcript/pyproject.toml
@@ -1,7 +1,7 @@
 [project]
 name = "audio_transcript"
 version = "0.0.1"
-description = "Transcribe audio with OpenAI Whisper and translate the text with OpenAI GPT-4o-mini"
+description = "Transcribe audio with OpenAI Whisper and translate the text with OpenAI gpt-4.1-mini"
 authors = [{ name = "Restack Team", email = "service@restack.io" }]
 requires-python = ">=3.10,<3.14"
 readme = "README.md"
diff --git a/audio_transcript/src/functions/translate_text.py b/audio_transcript/src/functions/translate_text.py
@@ -20,7 +20,7 @@ async def translate_text(input: TranslateTextInput):
 
     try:
       response = client.chat.completions.create(
-          model="gpt-4o-mini",
+          model="gpt-4.1-mini",
           messages=[
               {
                   "role": "system",
diff --git a/community/defense_quickstart_audio_transcription_translation/src/functions/translate.py b/community/defense_quickstart_audio_transcription_translation/src/functions/translate.py
@@ -23,7 +23,7 @@ async def translate(input: FunctionInputParams):
         print(messages)
         messages.append({"role": "system", "content": "To each output in the end add a line 'Helped By Restack AI'"})
         response = client.chat.completions.create(
-            model="gpt-4o-mini",
+            model="gpt-4.1-mini",
             messages=messages,
             temperature=0.0
         )
diff --git a/community/e2b/src/workflows/code_execution.py b/community/e2b/src/workflows/code_execution.py
@@ -21,7 +21,7 @@ async def run(self, input: CodeExecutionWorkflowInput) -> CodeExecutionWorkflowO
         messages = []
         while True:
             llm_response = await workflow.step(openai_tool_call, input=OpenaiToolCallInput(
-                model="gpt-4o-mini",
+                model="gpt-4.1-mini",
                 user_content=input.user_content if not messages else None,
                 system_content=input.system_content if not messages else None,
                 messages=messages,
diff --git a/community/livekit_opentelemetry/src/pipeline.py b/community/livekit_opentelemetry/src/pipeline.py
@@ -112,7 +112,7 @@ async def entrypoint(ctx: JobContext) -> None:
         vad=ctx.proc.userdata["vad"],
         stt=deepgram.STT(),
         llm=openai.LLM(
-            # model="gpt-4o-mini",
+            # model="gpt-4.1-mini",
             # api_key=os.environ.get("OPENAI_API_KEY"),
             api_key=f"{agent_id}-livekit",
             base_url=agent_url,
diff --git a/community/openai_greet/src/functions/function.py b/community/openai_greet/src/functions/function.py
@@ -24,7 +24,7 @@ async def openai_greet(input: FunctionInputParams) -> str:
         messages.append({"role": "user", "content": input.user_content})
 
         response = client.chat.completions.create(
-            model=input.model or "gpt-4o-mini",
+            model=input.model or "gpt-4.1-mini",
             messages=messages,
             response_format={
                 "json_schema": {
diff --git a/community/re_act/src/functions/decide.py b/community/re_act/src/functions/decide.py
@@ -39,7 +39,7 @@ async def decide(input: DecideInput):
         ]
 
         response = client.chat.completions.create(
-            model="gpt-4o-mini",
+            model="gpt-4.1-mini",
             messages=[
                 {
                     "role": "system",
diff --git a/community/re_act/src/functions/generate_email_content.py b/community/re_act/src/functions/generate_email_content.py
@@ -17,7 +17,7 @@ async def generate_email_content(input: GenerateEmailInput):
         client = OpenAI(api_key=os.environ.get("OPENAI_API_KEY"))
 
         response = client.chat.completions.create(
-            model="gpt-4o-mini",
+            model="gpt-4.1-mini",
             messages=[
                 {
                     "role": "system",
diff --git a/pdf_ocr/src/functions/openai_chat.py b/pdf_ocr/src/functions/openai_chat.py
@@ -28,7 +28,7 @@ async def openai_chat(input: OpenAiChatInput) -> str:
         messages.append({"role": "user", "content": input.user_content})
 
         response = client.chat.completions.create(
-            model=input.model or "gpt-4o-mini",
+            model=input.model or "gpt-4.1-mini",
             messages=messages
         )
         log.info("openai_chat function completed", response=response)
diff --git a/pdf_ocr/src/workflows/pdf.py b/pdf_ocr/src/workflows/pdf.py
@@ -34,7 +34,7 @@ async def run(self, input: PdfWorkflowInput):
                     function=openai_chat,
                     function_input=OpenAiChatInput(
                         user_content=f"Make a summary of that PDF. Here is the OCR result: {ocr_result}",
-                        model="gpt-4o-mini"
+                        model="gpt-4.1-mini"
                     ),
                     start_to_close_timeout=timedelta(seconds=120)
                 )

Original file line number	Diff line number	Diff line change
`@@ -41,7 +41,7 @@ async def llm(function_input: FunctionInputParams) -> str:`
`41`	`41`	`messages.append({"role": "user", "content": function_input.user_content})`
`42`	`42`
`43`	`43`	`response = client.chat.completions.create(`
`44`		`- model=function_input.model or "gpt-4o-mini", messages=messages`
	`44`	`+ model=function_input.model or "gpt-4.1-mini", messages=messages`
`45`	`45`	`)`
`46`	`46`	`log.info("llm function completed", response=response)`
`47`	`47`	`return response.choices[0].message.content`
Original file line number	Diff line number	Diff line change
`@@ -35,7 +35,7 @@ async def run(self, workflow_input: WorkflowInputParams) -> dict:`
`35`	`35`	`function_input=FunctionInputParams(`
`36`	`36`	`system_content=f"You are a personal assitant and have access to weather data {weather_data}. Always greet person with relevant info from weather data",`
`37`	`37`	`user_content=user_content,`
`38`		`- model="gpt-4o-mini",`
	`38`	`+ model="gpt-4.1-mini",`
`39`	`39`	`),`
`40`	`40`	`start_to_close_timeout=timedelta(seconds=120),`
`41`	`41`	`)`
Original file line number	Diff line number	Diff line change
`@@ -44,7 +44,7 @@ async def llm_chat(agent_input: LlmChatInput) -> dict[str, str]:`
`44`	`44`	`)`
`45`	`45`
`46`	`46`	`assistant_raw_response = client.chat.completions.create(`
`47`		`- model=agent_input.model or "gpt-4o-mini",`
	`47`	`+ model=agent_input.model or "gpt-4.1-mini",`
`48`	`48`	`messages=agent_input.messages,`
`49`	`49`	`)`
`50`	`50`	`except Exception as e:`
Original file line number	Diff line number	Diff line change
`@@ -45,7 +45,7 @@ async def llm_chat(function_input: LlmChatInput) -> ChatCompletion:`
`45`	`45`	`)`
`46`	`46`
`47`	`47`	`response = client.chat.completions.create(`
`48`		`- model=function_input.model or "gpt-4o-mini",`
	`48`	`+ model=function_input.model or "gpt-4.1-mini",`
`49`	`49`	`messages=function_input.messages,`
`50`	`50`	`)`
`51`	`51`	`except Exception as e:`
Original file line number	Diff line number	Diff line change
`@@ -40,7 +40,7 @@ async def llm_chat(function_input: LlmChatInput) -> str:`
`40`	`40`	`messages_dicts = [message.model_dump() for message in function_input.messages]`
`41`	`41`	`# Get the streamed response from OpenAI API`
`42`	`42`	`response: Stream[ChatCompletionChunk] = client.chat.completions.create(`
`43`		`- model=function_input.model or "gpt-4o-mini",`
	`43`	`+ model=function_input.model or "gpt-4.1-mini",`
`44`	`44`	`messages=messages_dicts,`
`45`	`45`	`stream=True,`
`46`	`46`	`)`
Original file line number	Diff line number	Diff line change
`@@ -55,7 +55,7 @@ async def llm_chat(function_input: LlmChatInput) -> ChatCompletion:`
`55`	`55`	`)`
`56`	`56`
`57`	`57`	`response = client.chat.completions.create(`
`58`		`- model=function_input.model or "gpt-4o-mini",`
	`58`	`+ model=function_input.model or "gpt-4.1-mini",`
`59`	`59`	`messages=function_input.messages,`
`60`	`60`	`tools=function_input.tools,`
`61`	`61`	`)`