1movmedia/pyworker
2
0
Fork 0
Code Pull Requests Activity

Use PyWorker SDK (#67)

Browse Source 
* Change PyWorker to Worker SDK
* Moved /lib to vast-sdk (https://github.com/vast-ai/vast-sdk)
This commit is contained in:
LucasArmandVast
2025-12-15 22:33:03 -05:00
committed by GitHub
parent 2ce741a8b7
commit 4380d98c01
54 changed files with 1622 additions and 4626 deletions
Download Patch File Download Diff File Expand all files Collapse all files
workers/openai/client.py
+27 -35
View File
@@ -102,15 +102,13 @@ async def call_completions(client: Serverless, *, model: str, prompt: str, endpo
endpoint = await client.get_endpoint(name=endpoint_name)
payload = {
"input": {
"model": model,
"prompt": prompt,
"max_tokens": kwargs.get("max_tokens", MAX_TOKENS),
"temperature": kwargs.get("temperature", DEFAULT_TEMPERATURE),
}
"model": model,
"prompt": prompt,
"max_tokens": kwargs.get("max_tokens", MAX_TOKENS),
"temperature": kwargs.get("temperature", DEFAULT_TEMPERATURE),
}
log.debug("POST /v1/completions %s", json.dumps(payload)[:500])
resp = await endpoint.request("/v1/completions", payload, cost=payload["input"]["max_tokens"])
resp = await endpoint.request("/v1/completions", payload, cost=payload["max_tokens"])
return resp["response"]
async def call_chat_completions(client: Serverless, *, model: str, messages: List[Dict[str, Any]], endpoint_name: str, **kwargs) -> Dict[str, Any]:
@@ -118,17 +116,15 @@ async def call_chat_completions(client: Serverless, *, model: str, messages: Lis
endpoint = await client.get_endpoint(name=endpoint_name)
payload = {
"input": {
"model": model,
"messages": messages,
"max_tokens": kwargs.get("max_tokens", MAX_TOKENS),
"temperature": kwargs.get("temperature", DEFAULT_TEMPERATURE),
**({"tools": kwargs["tools"]} if "tools" in kwargs else {}),
**({"tool_choice": kwargs["tool_choice"]} if "tool_choice" in kwargs else {}),
}
"model": model,
"messages": messages,
"max_tokens": kwargs.get("max_tokens", MAX_TOKENS),
"temperature": kwargs.get("temperature", DEFAULT_TEMPERATURE),
**({"tools": kwargs["tools"]} if "tools" in kwargs else {}),
**({"tool_choice": kwargs["tool_choice"]} if "tool_choice" in kwargs else {}),
}
log.debug("POST /v1/chat/completions %s", json.dumps(payload)[:500])
resp = await endpoint.request("/v1/chat/completions", payload, cost=payload["input"]["max_tokens"])
resp = await endpoint.request("/v1/chat/completions", payload, cost=payload["max_tokens"])
return resp["response"]
# ---- Streaming variants ----
@@ -137,17 +133,15 @@ async def stream_completions(client: Serverless, *, model: str, prompt: str, end
endpoint = await client.get_endpoint(name=endpoint_name)
payload = {
"input": {
"model": model,
"prompt": prompt,
"max_tokens": kwargs.get("max_tokens", MAX_TOKENS),
"temperature": kwargs.get("temperature", DEFAULT_TEMPERATURE),
"stream": True,
**({"stop": kwargs["stop"]} if "stop" in kwargs else {}),
}
"model": model,
"prompt": prompt,
"max_tokens": kwargs.get("max_tokens", MAX_TOKENS),
"temperature": kwargs.get("temperature", DEFAULT_TEMPERATURE),
"stream": True,
**({"stop": kwargs["stop"]} if "stop" in kwargs else {}),
}
log.debug("STREAM /v1/completions %s", json.dumps(payload)[:500])
resp = await endpoint.request("/v1/completions", payload, cost=payload["input"]["max_tokens"], stream=True)
resp = await endpoint.request("/v1/completions", payload, cost=payload["max_tokens"], stream=True)
return resp["response"] # async generator
async def stream_chat_completions(client: Serverless, *, model: str, messages: List[Dict[str, Any]], endpoint_name: str, **kwargs):
@@ -155,18 +149,16 @@ async def stream_chat_completions(client: Serverless, *, model: str, messages: L
endpoint = await client.get_endpoint(name=endpoint_name)
payload = {
"input": {
"model": model,
"messages": messages,
"max_tokens": kwargs.get("max_tokens", MAX_TOKENS),
"temperature": kwargs.get("temperature", DEFAULT_TEMPERATURE),
"stream": True,
**({"tools": kwargs["tools"]} if "tools" in kwargs else {}),
**({"tool_choice": kwargs["tool_choice"]} if "tool_choice" in kwargs else {}),
}
"model": model,
"messages": messages,
"max_tokens": kwargs.get("max_tokens", MAX_TOKENS),
"temperature": kwargs.get("temperature", DEFAULT_TEMPERATURE),
"stream": True,
**({"tools": kwargs["tools"]} if "tools" in kwargs else {}),
**({"tool_choice": kwargs["tool_choice"]} if "tool_choice" in kwargs else {}),
}
log.debug("STREAM /v1/chat/completions %s", json.dumps(payload)[:500])
resp = await endpoint.request("/v1/chat/completions", payload, cost=payload["input"]["max_tokens"], stream=True)
resp = await endpoint.request("/v1/chat/completions", payload, cost=payload["max_tokens"], stream=True)
return resp["response"] # async generator