shekkari21 commited on
Commit
ac32153
·
1 Parent(s): ece54a0

changed base model

Browse files
Files changed (3) hide show
  1. __pycache__/config.cpython-311.pyc +0 -0
  2. config.py +3 -3
  3. test_local.sh +29 -0
__pycache__/config.cpython-311.pyc CHANGED
Binary files a/__pycache__/config.cpython-311.pyc and b/__pycache__/config.cpython-311.pyc differ
 
config.py CHANGED
@@ -17,8 +17,8 @@ OPENAI_API_KEY = os.getenv("OPENAI_API_KEY")
17
  OPENAI_MODEL = os.getenv("OPENAI_MODEL", "gpt-4o")
18
 
19
  # Ollama Configuration (for local open-source models)
20
- OLLAMA_BASE_URL = os.getenv("OLLAMA_BASE_URL", "http://localhost:11434")
21
- OLLAMA_MODEL = os.getenv("OLLAMA_MODEL", "llama3.2") # Options: llama3.2, mistral, qwen2.5, etc.
22
 
23
  # LiteLLM Configuration (for Hugging Face or other providers)
24
  LITELLM_MODEL = os.getenv("LITELLM_MODEL", "huggingface/meta-llama/Llama-3.2-3B-Instruct")
@@ -26,7 +26,7 @@ LITELLM_API_KEY = os.getenv("LITELLM_API_KEY", "") # Optional, depends on provi
26
 
27
  # Hugging Face Configuration (for using HF Inference API)
28
  HF_API_KEY = os.getenv("HF_API_KEY", "") # Get from https://huggingface.co/settings/tokens
29
- HF_MODEL = os.getenv("HF_MODEL", "meta-llama/Llama-3.1-8B-Instruct") # Best open-source model (8B params)
30
 
31
  # OpenRouter Configuration (backup option)
32
  OPENROUTER_API_KEY = os.getenv("OPENROUTER_API_KEY")
 
17
  OPENAI_MODEL = os.getenv("OPENAI_MODEL", "gpt-4o")
18
 
19
  # Ollama Configuration (for local open-source models)
20
+ OLLAMA_BASE_URL = os.getenv("OLLAMA_BASE_URL", "http://localhost:11434/v1") # /v1 for OpenAI-compatible API
21
+ OLLAMA_MODEL = os.getenv("OLLAMA_MODEL", "mistral") # Model name (Ollama handles :latest automatically)
22
 
23
  # LiteLLM Configuration (for Hugging Face or other providers)
24
  LITELLM_MODEL = os.getenv("LITELLM_MODEL", "huggingface/meta-llama/Llama-3.2-3B-Instruct")
 
26
 
27
  # Hugging Face Configuration (for using HF Inference API)
28
  HF_API_KEY = os.getenv("HF_API_KEY", "") # Get from https://huggingface.co/settings/tokens
29
+ HF_MODEL = os.getenv("HF_MODEL", "Qwen/Qwen2.5-7B-Instruct") # Best open-source model (8B params)
30
 
31
  # OpenRouter Configuration (backup option)
32
  OPENROUTER_API_KEY = os.getenv("OPENROUTER_API_KEY")
test_local.sh ADDED
@@ -0,0 +1,29 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ #!/bin/bash
2
+ # Local testing script with Ollama
3
+
4
+ echo "Setting up Ollama for local testing..."
5
+
6
+ # Set environment variables for Ollama
7
+ export LLM_PROVIDER=ollama
8
+ export OLLAMA_MODEL=mistral
9
+ export OLLAMA_BASE_URL=http://localhost:11434/v1
10
+
11
+ # Check if Ollama is running
12
+ if ! curl -s http://localhost:11434/api/tags > /dev/null 2>&1; then
13
+ echo "⚠️ Warning: Ollama service doesn't seem to be running!"
14
+ echo " Start it with: ollama serve"
15
+ echo ""
16
+ read -p "Continue anyway? (y/n) " -n 1 -r
17
+ echo
18
+ if [[ ! $REPLY =~ ^[Yy]$ ]]; then
19
+ exit 1
20
+ fi
21
+ fi
22
+
23
+ echo "✅ Using Ollama with model: $OLLAMA_MODEL"
24
+ echo ""
25
+
26
+ # Run the app
27
+ echo "Starting app..."
28
+ python app.py
29
+