phoenix/litellm-mirror
0
0
Fork 1
mirror of https://github.com/BerriAI/litellm.git synced 2025-04-27 11:43:54 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions
11619 commits 2497 branches 897 tags 550 MiB
Commit graph

2077 commits

Author SHA1 Message Date
Krrish Dholakia
0c1fc34e9f feat(slack_alerting.py): support sending daily reports on deployments 
allow admin to easily know slow + failing deployments

Closes  https://github.com/BerriAI/litellm/issues/3483
 2024-05-06 17:18:42 -07:00
Ishaan Jaff
881313f335 feat - send alert on adding new model 2024-05-06 15:45:07 -07:00
Ishaan Jaff
e5488a94dd fix - add better debugging on num_callbacks test 2024-05-06 13:42:20 -07:00
Ishaan Jaff
e1688d9d8b ui - new build 2024-05-04 10:55:39 -07:00
Lunik
5f43a7b511 🔊 fix: Correctly use verbose logging 
Signed-off-by: Lunik <lunik@tiwabbit.fr>
 2024-05-04 11:04:23 +02:00
Lunik
38d4cbc511 ✨ feat: Use 8 severity levels for azure content safety 
Signed-off-by: Lunik <lunik@tiwabbit.fr>
 2024-05-04 10:45:39 +02:00
Lunik
d69a1eeb4f 📝 doc: Azure content safety Proxy usage 
Signed-off-by: Lunik <lunik@tiwabbit.fr>
 2024-05-04 10:39:43 +02:00
Ishaan Jaff
5ffe9a8008 fix - select startTime and endTime on UI 2024-05-03 21:20:19 -07:00
Ishaan Jaff
1f0f0bcfc5 ui - new build 2024-05-03 20:40:58 -07:00
Marc Abramowitz
15ba2977ef Add route: "/engines/{model:path}/chat/completions" 
Without this, it results in:

```pytb
Traceback (most recent call last):
  File "/Users/abramowi/Code/OpenSource/litellm/litellm/proxy/proxy_server.py", line 3836, in completion
    raise HTTPException(
fastapi.exceptions.HTTPException: 400: {'error': 'completion: Invalid model name passed in model=gpt-3.5-turbo/chat'}
```
 2024-05-03 18:02:29 -07:00
Ishaan Jaff
ca054f95de feat - filter exceptions by model group 2024-05-03 16:54:24 -07:00
Ishaan Jaff
a0136af9ec Merge pull request #3427 from BerriAI/litellm_test_alert_size 
[Test] - Ensure only 1 slack callback + Size of of all callbacks do not grow
 2024-05-03 16:27:16 -07:00
Krish Dholakia
53993c612b Merge pull request #3430 from BerriAI/litellm_return_api_base 
feat(proxy_server.py): return api base in response headers
 2024-05-03 16:25:21 -07:00
Krrish Dholakia
aca36d8b79 feat(proxy_server.py): return api base in response headers 
Closes https://github.com/BerriAI/litellm/issues/2631
 2024-05-03 15:27:32 -07:00
Ishaan Jaff
af9875f358 fix test slack alerting len 2024-05-03 14:58:11 -07:00
Ishaan Jaff
5b97d56e41 fix - return num callbacks in /active/callbacks 2024-05-03 14:24:01 -07:00
Ishaan Jaff
2c62353ce0 Merge pull request #3426 from BerriAI/litellm_set_db_exceptions_on_ui 
UI - set DB Exceptions webhook_url on UI
 2024-05-03 14:05:37 -07:00
Ishaan Jaff
b334b60c41 fix bug where slack would get inserting several times 2024-05-03 14:04:38 -07:00
Lunik
08593fcaab ⚡️ perf: Remove test violation on each stream chunk 
Signed-off-by: Lunik <lunik@tiwabbit.fr>
 2024-05-03 20:51:40 +02:00
Lunik
7945e28356 ✅ ci: Add tests 
Signed-off-by: Lunik <lunik@tiwabbit.fr>
 2024-05-03 20:50:37 +02:00
Ishaan Jaff
051d4daa08 proxy - return num callbacks on /health/readiness 2024-05-03 09:14:32 -07:00
Krrish Dholakia
4a22948ce4 fix(proxy/utils.py): fix slack alerting to only raise alerts for llm api exceptions 
don't spam for bad user requests. Closes https://github.com/BerriAI/litellm/issues/3395
 2024-05-02 17:18:21 -07:00
Marc Abramowitz
24ddba804f Disambiguate invalid model name errors 
because that error can be thrown in several different places, so
knowing the function it's being thrown from can be very useul for debugging.
 2024-05-02 15:02:54 -07:00
Krrish Dholakia
c1ffa7d409 fix(proxy/utils.py): fix retry logic for generic data request 2024-05-02 14:50:50 -07:00
Lunik
3ca174bc57 ✨ feat: Add Azure Content-Safety Proxy hooks 
Signed-off-by: Lunik <lunik@tiwabbit.fr>
 2024-05-02 23:21:08 +02:00
Krish Dholakia
1b5efc47e1 Merge pull request #3375 from msabramo/GH-3372 
Fix route `/openai/deployments/{model}/chat/completions` not working properly
 2024-05-02 13:00:25 -07:00
Krrish Dholakia
588751ec86 refactor(main.py): trigger new build 2024-05-01 21:59:33 -07:00
Ishaan Jaff
a8d888171a ui - new build 2024-05-01 21:43:00 -07:00
Krish Dholakia
dff4649509 Merge branch 'main' into litellm_openmeter_integration 2024-05-01 21:19:29 -07:00
Krrish Dholakia
0b1382aef4 build(ui): enable adding openmeter via proxy ui 2024-05-01 21:16:23 -07:00
Ishaan Jaff
ed9fd1326c ui - new build 2024-05-01 19:52:57 -07:00
Krrish Dholakia
e7b3ac8e06 feat(openmeter.py): add support for user billing 
open-meter supports user based billing. Closes https://github.com/BerriAI/litellm/issues/1268
 2024-05-01 17:23:48 -07:00
Ishaan Jaff
5ce76a5b0d feat - show slow count and total count 2024-05-01 17:18:14 -07:00
Ishaan Jaff
1721ab64aa feat - return slow responses on admin UI 2024-05-01 17:16:33 -07:00
Ishaan Jaff
cb5073da62 fix - dont show cache hits on model latency tracker 2024-05-01 16:51:15 -07:00
Ishaan Jaff
5b70130049 fix latency tracking tool tip 2024-05-01 16:47:30 -07:00
Ishaan Jaff
c99df93515 ui - new build 2024-05-01 13:32:32 -07:00
Ishaan Jaff
9808bb8509 Merge pull request #3380 from BerriAI/ui_polish_viewing_model_latencies 
[UI] Polish viewing Model Latencies
 2024-05-01 09:44:53 -07:00
Ishaan Jaff
cf2dd97f0b clean up model latency metrics 2024-05-01 08:27:01 -07:00
Krrish Dholakia
0b2c3f78d4 fix(proxy/utils.py): emit number of spend transactions for keys being written to db in a batch 2024-05-01 08:25:04 -07:00
Ishaan Jaff
26ffb6c4a3 fix - prisma schema 2024-04-30 23:09:53 -07:00
Ishaan Jaff
8a6b66fe1f (ui - new build) 2024-04-30 22:54:51 -07:00
Ishaan Jaff
1a2045d63a ui - show tokens / sec 2024-04-30 22:44:28 -07:00
Ishaan Jaff
ee5ebd8f4f fix - viewing model metrics 2024-04-30 18:26:14 -07:00
Ishaan Jaff
69d8f7b89b fix - track litellm_model_name in LiteLLM_ErrorLogs 2024-04-30 17:31:40 -07:00
Ishaan Jaff
5756adc178 backend - show model latency per token 2024-04-30 17:23:36 -07:00
Ishaan Jaff
1c9aa7f0f1 fix passing starttime and endtime to model/exceptions 2024-04-30 16:53:53 -07:00
Ishaan Jaff
672d1c67fc fix - backend return exceptions 2024-04-30 15:41:16 -07:00
Marc Abramowitz
eac4593ab9 Move chat_completions before completions 
so that the `chat_completions` route is defined before the `completions` route.
This is necessary because the `chat_completions` route is more
specific than the `completions` route, and the order of route definitions
matters in FastAPI.

Without this, doing a request to
`/openai/deployments/{model_in_url}/chat/completions` might trigger
`completions` being called (with `model` set to `{model_in_url}/chat` instead of
`chat_completions` getting called, which is the correct function.

Fixes: GH-3372
 2024-04-30 15:07:10 -07:00
Ishaan Jaff
1776f5aea4 stash /model/metrics/exceptions endpoints 2024-04-30 14:19:23 -07:00