admtracksteel/BotVPS
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Change default Ollama model from qwen2.5-coder to llama3.2:1b for faster chat

Browse Source
This commit is contained in:
Marcos
2026-03-22 17:25:11 -03:00
parent 17dcb9d178
commit 8002262cf7
2 changed files with 6 additions and 6 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
ai_agent.py
View File

@@ -24,7 +24,7 @@ def get_llm_response(prompt: str, provider: str, cfg: dict) -> str:
elif provider == "ollama": elif provider == "ollama":
ollama_host = os.getenv("OLLAMA_HOST", "http://ollama:11434") ollama_host = os.getenv("OLLAMA_HOST", "http://ollama:11434")
model = os.getenv("OLLAMA_MODEL", "qwen2.5-coder:1.5b") model = os.getenv("OLLAMA_MODEL", "llama3.2:1b")
try: try:
res = requests.post(f"{ollama_host}/api/generate", json={ res = requests.post(f"{ollama_host}/api/generate", json={
"model": model, "model": model,

10
llm_providers.py
View File

@@ -39,7 +39,7 @@ LLM_PROVIDERS = {
"type": "local", "type": "local",
"endpoint": os.getenv("OLLAMA_HOST", "http://ollama:11434"), "endpoint": os.getenv("OLLAMA_HOST", "http://ollama:11434"),
"models": None, "models": None,
"default": "qwen2.5-coder:1.5b" "default": "llama3.2:1b"
} }
} }
@@ -64,7 +64,7 @@ def get_config() -> dict:
return { return {
"orchestrator": { "orchestrator": {
"planner": {"provider": "gemini", "model": "gemini-2.5-flash"}, "planner": {"provider": "gemini", "model": "gemini-2.5-flash"},
"executor": {"provider": "ollama", "model": "qwen2.5-coder:1.5b"} "executor": {"provider": "ollama", "model": "llama3.2:1b"}
}, },
"api_keys": { "api_keys": {
"openai": "", "openai": "",
@@ -85,7 +85,7 @@ def get_orchestrator_config() -> dict:
cfg = get_config() cfg = get_config()
return cfg.get("orchestrator", { return cfg.get("orchestrator", {
"planner": {"provider": "gemini", "model": "gemini-2.5-flash"}, "planner": {"provider": "gemini", "model": "gemini-2.5-flash"},
"executor": {"provider": "ollama", "model": "qwen2.5-coder:1.5b"} "executor": {"provider": "ollama", "model": "llama3.2:1b"}
}) })
def set_planner(provider: str = None, model: str = None) -> dict: def set_planner(provider: str = None, model: str = None) -> dict:
@@ -116,7 +116,7 @@ def set_executor(provider: str = None, model: str = None) -> dict:
} }
save_config(cfg) save_config(cfg)
return cfg["orchestrator"].get("executor", {"provider": "ollama", "model": "qwen2.5-coder:1.5b"}) return cfg["orchestrator"].get("executor", {"provider": "ollama", "model": "llama3.2:1b"})
return cfg["orchestrator"]["executor"] return cfg["orchestrator"]["executor"]
def set_api_key(provider: str, key: str): def set_api_key(provider: str, key: str):
@@ -370,7 +370,7 @@ def get_planner_llm() -> tuple:
def get_executor_llm() -> tuple: def get_executor_llm() -> tuple:
"""Retorna provider e modelo do executor configurado.""" """Retorna provider e modelo do executor configurado."""
cfg = get_orchestrator_config() cfg = get_orchestrator_config()
executor = cfg.get("executor", {"provider": "ollama", "model": "qwen2.5-coder:1.5b"}) executor = cfg.get("executor", {"provider": "ollama", "model": "llama3.2:1b"})
return executor["provider"], executor["model"] return executor["provider"], executor["model"]
def call_planner(prompt: str, system_prompt: str = None) -> str: def call_planner(prompt: str, system_prompt: str = None) -> str: