Spaces:
Running
Running
ffreemt
commited on
Commit
·
88aceaf
1
Parent(s):
c5a9402
Update model_group_alias
Browse files- .gitignore +1 -0
- config.yaml +6 -2
.gitignore
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
.env
|
config.yaml
CHANGED
|
@@ -73,10 +73,14 @@ litellm_settings:
|
|
| 73 |
# Networking settings
|
| 74 |
request_timeout: 20 # (int) llm request timeout in seconds. Raise Timeout error if call takes longer than 10s. Sets litellm.request_timeout
|
| 75 |
num_retries: 3
|
| 76 |
-
fallbacks: [{"gemini-1.5-pro": ["
|
| 77 |
allowed_fails: 3 # cooldown model if it fails > 1 call in a minute.
|
| 78 |
cooldown_time: 30 # how long to cooldown model if fails/min > allowed_fails
|
| 79 |
|
| 80 |
general_settings:
|
| 81 |
master_key: os.environ/MASTER_KEY # sk-1234 # [OPTIONAL] Only use this if you require all calls to contain this key (Authorization: Bearer sk-1234)
|
| 82 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 73 |
# Networking settings
|
| 74 |
request_timeout: 20 # (int) llm request timeout in seconds. Raise Timeout error if call takes longer than 10s. Sets litellm.request_timeout
|
| 75 |
num_retries: 3
|
| 76 |
+
fallbacks: [{"gemini-1.5-pro": ["gemini-1.5-flash"]}]
|
| 77 |
allowed_fails: 3 # cooldown model if it fails > 1 call in a minute.
|
| 78 |
cooldown_time: 30 # how long to cooldown model if fails/min > allowed_fails
|
| 79 |
|
| 80 |
general_settings:
|
| 81 |
master_key: os.environ/MASTER_KEY # sk-1234 # [OPTIONAL] Only use this if you require all calls to contain this key (Authorization: Bearer sk-1234)
|
| 82 |
+
|
| 83 |
+
router_settings:
|
| 84 |
+
fallbacks: [{"gpt-4o-mini": ["gemini-1.5-flash"]}]
|
| 85 |
+
model_group_alias: {"gpt-4": "gemini-1.5-pro"}
|
| 86 |
+
routing_strategy: simple-shuffle
|