Ishaan Jaff
37069f45cb
fix - use one slack class
2024-04-23 18:31:15 -07:00
Krrish Dholakia
161e836427
fix(utils.py): fix 'no-cache': true when caching is turned on
2024-04-23 12:58:30 -07:00
Ishaan Jaff
5118c96b4d
fix langfuse project id
2024-04-23 11:37:17 -07:00
Krrish Dholakia
6ac0dba5c2
fix(prometheus.py): add user tracking to prometheus
2024-04-22 15:14:38 -07:00
Krrish Dholakia
9379e3d047
fix(lowest_tpm_rpm_v2.py): use a combined tpm+rpm query in async get cache, to reduce redis client calls in high traffic
2024-04-20 16:13:11 -07:00
Krish Dholakia
6d9f0f1839
Merge branch 'main' into litellm_ssl_caching_fix
2024-04-19 17:20:27 -07:00
Krish Dholakia
cc7f27252b
Merge pull request #3179 from BerriAI/litellm_dual_caching_fix
...
fix(caching.py): dual cache async_batch_get_cache fix + testing
2024-04-19 17:02:47 -07:00
Ishaan Jaff
788394b1ec
fix - logging proxy base url to langfuse
2024-04-19 16:20:46 -07:00
Krrish Dholakia
0d9c96bebf
test(test_prometheus_services.py): fix testing to handle caching ping in init
2024-04-19 16:15:29 -07:00
Ishaan Jaff
c59051895b
fix - allow users to opt into langfuse default tags
2024-04-19 16:05:31 -07:00
Ishaan Jaff
c0befe9bbb
feat - log base_url to langfuse as a tag
2024-04-19 15:48:00 -07:00
Ishaan Jaff
48e5de4f0e
Merge pull request #3175 from BerriAI/litellm_log_team_alias_langfuse
...
[FEAT] Log team alias to langfuse
2024-04-19 15:30:59 -07:00
Krrish Dholakia
01a1a8f731
fix(caching.py): dual cache async_batch_get_cache fix + testing
...
this fixes a bug in usage-based-routing-v2 which was caused b/c of how the result was being returned from dual cache async_batch_get_cache. it also adds unit testing for that function (and it's sync equivalent)
2024-04-19 15:03:25 -07:00
Ishaan Jaff
3167c9da9f
fix - use user_api_key_team_alias as the default tag on langfuse
2024-04-19 14:43:41 -07:00
Ishaan Jaff
410d1f2d2c
langfuse - log team alias
2024-04-19 14:26:58 -07:00
Krrish Dholakia
08cf77623b
fix(caching.py): remove url parsing logic - causing redis ssl connections to fail
...
this reverts a change that was causing redis url w/ ssl to fail. this also adds unit testing for this sc
enario, to prevent future regressions
2024-04-19 14:01:13 -07:00
Ishaan Jaff
b2bdc99474
Merge pull request #2961 from Manouchehri/langfuse-add-region-2937
...
(feat) Langfuse - Add location logging, and add cache_hit to metadata.
2024-04-19 10:06:34 -07:00
Krrish Dholakia
81573b2dd9
fix(test_lowest_tpm_rpm_routing_v2.py): unit testing for usage-based-routing-v2
2024-04-18 21:38:00 -07:00
Krrish Dholakia
0f95a824c4
feat(prometheus_services.py): emit proxy latency for successful llm api requests
...
uses prometheus histogram for this
2024-04-18 16:04:35 -07:00
Krrish Dholakia
d8d78bfb4f
fix(prometheus.py): fix metric name to be more accurate
...
change metric name from litellm_failed_requests_metric to litellm_llm_api_failed_requests_metric
2024-04-18 12:30:44 -07:00
Krrish Dholakia
28edb77350
fix(utils.py): support prometheus failed call metrics
2024-04-18 12:29:15 -07:00
greenscale-nandesh
907e3973fd
Merge branch 'BerriAI:main' into main
2024-04-17 12:24:29 -07:00
Ishaan Jaff
53388b4436
fix don't log user_api_key to prometheus
2024-04-16 19:01:38 -07:00
greenscale-nandesh
3feb0ef897
Merge branch 'BerriAI:main' into main
2024-04-16 11:49:26 -07:00
Vivek Aditya
9c3a1124d7
Athina callback now handles streaming mode
2024-04-17 00:06:16 +05:30
David Manouchehri
ce13d2fa7c
(feat) - Improve Langfuse to log request headers.
...
https://github.com/orgs/langfuse/discussions/1070
2024-04-16 17:38:25 +00:00
David Manouchehri
af9da303b3
(feat) Langfuse - Add location logging, and add cache_hit to metadata.
2024-04-16 17:37:24 +00:00
Krish Dholakia
9ab6482466
Merge pull request #2883 from Manouchehri/fix-broke-langfuse-2857
...
(fix) Langfuse v2 renamed a few things.
2024-04-15 16:14:05 -07:00
Krrish Dholakia
67d0d5e356
fix(prometheus_services.py): add better import error statement
2024-04-13 19:03:32 -07:00
Krrish Dholakia
bef24cd4ab
fix(caching.py): don't decode a string
2024-04-13 18:48:03 -07:00
Krrish Dholakia
9f42d15713
feat(prometheus_services.py): track when redis calls fail
2024-04-13 18:31:35 -07:00
Krrish Dholakia
4e81acf2c6
feat(prometheus_services.py): monitor health of proxy adjacent services (redis / postgres / etc.)
2024-04-13 18:15:02 -07:00
Krrish Dholakia
3b803b09d9
fix(langfuse.py): remove print statements
2024-04-11 17:23:41 -07:00
garfeildma
6aaef90ac9
fix(integrations/langsmith.py): fix datetime json serialize error
2024-04-09 11:19:26 +08:00
David Manouchehri
8682573187
(fix) Langfuse v2 renamed a few things.
2024-04-07 04:31:25 +00:00
Nandesh Guru
0e9b1f5247
Greenscale Integration
...
Adding logger for Greenscale
2024-04-04 15:38:51 -07:00
Nicolas van Kempen
4270337092
Use packaging over deprecated pkg_resources
...
Fix #2818
2024-04-04 14:08:15 -04:00
Krrish Dholakia
24d9fcb32c
fix(lunary.py): move parse_version to inside lunarylogger
2024-04-03 13:52:42 -07:00
Krrish Dholakia
fcaa452ccd
feat(prometheus.py): track team based metrics on prometheus
2024-04-03 13:43:21 -07:00
Krrish Dholakia
f07500c5ea
fix(proxy_server.py): bump default db timeouts
2024-04-03 09:35:08 -07:00
Krrish Dholakia
0d949d71ab
fix(main.py): support text completion input being a list of strings
...
addresses - https://github.com/BerriAI/litellm/issues/2792 , https://github.com/BerriAI/litellm/issues/2777
2024-04-02 08:50:16 -07:00
Vincelwt
1b84dfac91
Merge branch 'main' into main
2024-03-30 13:21:53 +09:00
Andrea Ponti
25bacfacdc
Add trace_name in langfuse logging v2
2024-03-27 16:58:08 +01:00
Krrish Dholakia
e10eb8f6fe
feat(llm_guard.py): enable key-specific llm guard check
2024-03-26 17:21:51 -07:00
Marc Klingen
e0ee94d45a
conditionally set sdk_integration
2024-03-24 16:46:31 +01:00
Vivek Aditya
6bd49c6087
Athina docs updated with information about additional fields and a minor fix in the callback
2024-03-23 12:42:07 +05:30
Vincelwt
29e8c144fb
Merge branch 'main' into main
2024-03-22 00:52:42 +09:00
Krrish Dholakia
d91f9a9f50
feat(proxy_server.py): enable llm api based prompt injection checks
...
run user calls through an llm api to check for prompt injection attacks. This happens in parallel to th
e actual llm call using `async_moderation_hook`
2024-03-20 22:43:42 -07:00
Ishaan Jaff
c196186190
(fix) add /metrics to utils.py
2024-03-19 17:28:33 -07:00
Ishaan Jaff
d2522752a6
(feat) prometheus metric collector
2024-03-19 15:44:36 -07:00
