apunkt/plano
1
0
Fork 0
mirror of https://github.com/katanemo/plano.git synced 2026-06-17 15:25:17 +02:00
Code Issues Projects Releases Packages Wiki Activity

refactor

Browse source
This commit is contained in:
Adil Hafeez 2025-03-14 17:34:18 -07:00
parent f4462d1ed5
commit 816f513663
No known key found for this signature in database
GPG key ID: 9B18EF7691369645
5 changed files with 2 additions and 20 deletions
Download patch file Download diff file Expand all files Collapse all files

15
crates/llm_gateway/src/stream_context.rs
View file

@ -214,7 +214,6 @@ impl HttpContext for StreamContext {
}));
} else {
self.select_llm_provider();
debug!("setting routing header to: {}", self.llm_provider().name);
self.add_http_request_header(
ARCH_ROUTING_HEADER,
&self.llm_provider().provider_interface.to_string(),
@ -243,12 +242,6 @@ impl HttpContext for StreamContext {
fn on_http_request_body(&mut self, body_size: usize, end_of_stream: bool) -> Action {
// Let the client send the gateway all the data before sending to the LLM_provider.
// TODO: consider a streaming API.
trace!(
"on_http_request_body [S={}] bytes={} end_stream={}",
self.context_id,
body_size,
end_of_stream
);
if self.request_body_sent_time.is_none() {
self.request_body_sent_time = Some(current_time_ns());
@ -316,14 +309,6 @@ impl HttpContext for StreamContext {
deserialized_body.model = model_name.to_string();
// if use_agent_orchestrator || self.llm_provider.as_ref().unwrap().model.is_none() {
// deserialized_body.model = "None".to_string()
// } else {
// // override model name from the llm provider
// deserialized_body
// .model
// .clone_from(&self.llm_provider.as_ref().unwrap().model.as_ref().unwrap());
// }
let chat_completion_request_str = serde_json::to_string(&deserialized_body).unwrap();
trace!(

2
crates/prompt_gateway/src/stream_context.rs
View file

@ -321,7 +321,7 @@ impl StreamContext {
metadata.insert("use_agent_orchestrator".to_string(), "true".to_string());
metadata.insert(
"Agent-Name".to_string(),
"agent-name".to_string(),
callout_context
.prompt_target_name
.as_ref()

2
demos/use_cases/orchestrating_agents/main.py
View file

@ -119,7 +119,7 @@ async def completion_api(req: ChatCompletionsRequest):
logger.info(f"request: {req}")
if req.metadata is None:
req.metadata = {}
agent_name = req.metadata.get("Agent-Name", "unknown agent")
agent_name = req.metadata.get("agent-name", "unknown agent")
logger.info(f"agent: {agent_name}")
agent_role = agent_map.get(agent_name)["role"]

2
model_server/src/commons/globals.py
View file

@ -21,13 +21,11 @@ ARCH_ENDPOINT = os.getenv("ARCH_ENDPOINT", "https://archfc.katanemo.dev/v1")
ARCH_API_KEY = "EMPTY"
ARCH_CLIENT = OpenAI(base_url=ARCH_ENDPOINT, api_key=ARCH_API_KEY)
ARCH_AGENT_CLIENT = ARCH_CLIENT
# ARCH_AGENT_CLIENT = OpenAI(api_key=os.getenv("OPENAI_API_KEY", "EMPTY"))
# Define model names
ARCH_INTENT_MODEL_ALIAS = "Arch-Intent"
ARCH_FUNCTION_MODEL_ALIAS = "Arch-Function"
ARCH_AGENT_MODEL_ALIAS = ARCH_FUNCTION_MODEL_ALIAS
# ARCH_AGENT_MODEL_ALIAS = "gpt-4o-mini"
ARCH_GUARD_MODEL_ALIAS = "katanemo/Arch-Guard"
# Define model handlers

1
model_server/src/core/function_calling.py
View file

@ -630,7 +630,6 @@ class ArchFunctionHandler(ArchBaseHandler):
return chat_completion_response
# override ArchFunctionHandler
class ArchAgentHandler(ArchFunctionHandler):
def __init__(self, client: OpenAI, model_name: str, config: ArchAgentConfig):
super().__init__(client, model_name, config)