Save cache_type to llm_response_cache
This commit is contained in:
@@ -568,6 +568,7 @@ async def save_to_cache(hashing_kv, cache_data: CacheData):
|
|||||||
|
|
||||||
mode_cache[cache_data.args_hash] = {
|
mode_cache[cache_data.args_hash] = {
|
||||||
"return": cache_data.content,
|
"return": cache_data.content,
|
||||||
|
"cache_type": cache_data.cache_data,
|
||||||
"embedding": cache_data.quantized.tobytes().hex()
|
"embedding": cache_data.quantized.tobytes().hex()
|
||||||
if cache_data.quantized is not None
|
if cache_data.quantized is not None
|
||||||
else None,
|
else None,
|
||||||
|
Reference in New Issue
Block a user