phoenix/litellm
1
0
Fork 0
forked from phoenix/litellm-mirror
Code Pull requests Activity

Merge pull request #2909 from CLARKBENHAM/clark/tpm_rpm_docs

Browse source 
Clark/tpm rpm docs
This commit is contained in:
Krish Dholakia 2024-04-09 21:12:30 -07:00 • committed by GitHub
commit 884f22c9d8
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
1 changed files with 12 additions and 4 deletions
Download patch file Download diff file Expand all files Collapse all files

16
docs/my-website/docs/routing.md
View file

@ -117,7 +117,10 @@ import asyncio
model_list = [{ ... }] model_list = [{ ... }]
# init router # init router
router = Router(model_list=model_list, routing_strategy="latency-based-routing") # 👈 set routing strategy router = Router(model_list=model_list,
routing_strategy="latency-based-routing",# 👈 set routing strategy
enable_pre_call_check=True, # enables router rate limits for concurrent calls
)
## CALL 1+2 ## CALL 1+2
tasks = [] tasks = []
@ -257,8 +260,9 @@ router = Router(model_list=model_list,
redis_host=os.environ["REDIS_HOST"], redis_host=os.environ["REDIS_HOST"],
redis_password=os.environ["REDIS_PASSWORD"], redis_password=os.environ["REDIS_PASSWORD"],
redis_port=os.environ["REDIS_PORT"], redis_port=os.environ["REDIS_PORT"],
routing_strategy="usage-based-routing") routing_strategy="usage-based-routing"
enable_pre_call_check=True, # enables router rate limits for concurrent calls
)
response = await router.acompletion(model="gpt-3.5-turbo", response = await router.acompletion(model="gpt-3.5-turbo",
messages=[{"role": "user", "content": "Hey, how's it going?"}] messages=[{"role": "user", "content": "Hey, how's it going?"}]
@ -555,7 +559,11 @@ router = Router(model_list: Optional[list] = None,
## Pre-Call Checks (Context Window) ## Pre-Call Checks (Context Window)
Enable pre-call checks to filter out deployments with context window limit < messages for a call. Enable pre-call checks to filter out:
1. deployments with context window limit < messages for a call.
2. deployments that have exceeded rate limits when making concurrent calls. (eg. `asyncio.gather(*[
router.acompletion(model="gpt-3.5-turbo", messages=m) for m in list_of_messages
])`)
<Tabs> <Tabs>
<TabItem value="sdk" label="SDK"> <TabItem value="sdk" label="SDK">