Refactor ask function in cli.py and completion_openai_gpt function in…

… core.py

- Refactor the ask function in cli.py to improve code readability and handle different models for completion.
- Refactor the completion_openai_gpt function in core.py to add support for streaming and handle different models for completion.

Update version in pyproject.toml

- Update the version in pyproject.toml from 0.3.2 to 0.3.3.

src/heygpt/cli.py

@@ -102,14 +102,26 @@ def ask(
         text = Prompt.ask("[blue]Enter text")
 
     # log.debug(text)
+    stream = False
+    if model.startswith("o1"):
+        completion = completion_openai_gpt(
+            command=command,
+            text=text,
+            model=model,
+            _print=raw,
+            stream=stream,
+        )
+    else:
+        stream = True
+        completion = completion_openai_gpt(
+            command=command,
+            text=text,
+            model=model,
+            _print=raw,
+            temperature=temperature,
+            stream=stream,
+        )
 
-    completion = completion_openai_gpt(
-        command=command,
-        text=text,
-        model=model,
-        _print=raw,
-        temperature=temperature,
-    )
     content = completion
 
     if not raw:

src/heygpt/core.py

@@ -43,7 +43,8 @@ def completion_openai_gpt(
     system: str = "",
     model=model,
     _print=False,
-    temperature=0.7,
+    temperature=None,
+    stream=True,
 ):
     """
     ref: https://docs.openai.com/api-reference/completions/create
@@ -66,21 +67,31 @@ def completion_openai_gpt(
     else:
         messages = [{"role": "user", "content": text}]
 
-    chat_completion = completion(
-        model=model,
-        messages=messages,
-        temperature=temperature,
-        stream=True,
-    )
-
-    for chunk in chat_completion:
-        # Process each chunk as needed
-        c = chunk.choices[0].delta.content or ""
-        out += c
-        if _print:
-            console.print(c, end="", markup=True)
-
-    return out
+    if stream:
+        chat_completion = completion(
+            model=model,
+            messages=messages,
+            temperature=temperature,
+            stream=True,
+        )
+
+        for chunk in chat_completion:
+            # Process each chunk as needed
+            c = chunk.choices[0].delta.content or ""
+            out += c
+            if _print:
+                console.print(c, end="", markup=True)
+
+        return out
+    else:
+        chat_completion = completion(
+            model=model,
+            messages=messages,
+            temperature=temperature,
+            stream=False,
+        )
+
+        return chat_completion.choices[0].message.content
 
 
 def wisper(audio_file):

src/heygpt/serve.py

@@ -40,7 +40,6 @@ def local_css(file_name="", style=""):
 
 # React to user input
 if chat_input := st.chat_input("What is up?"):
-
     # set stream to True if model is OpenAI
     stream = True
     if user_model.startswith("o1-"):

src/pyproject.toml

@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "heygptcli"
-version = "0.3.2"
+version = "0.3.3"
 description = ""
 authors = ["Rishang <rishang@localhost.com>"]
 readme = "README.md"