Ishaan Jaffer
3df0fb0b45
test fix
2025-10-17 10:46:42 -07:00
Ishaan Jaff
3b1fdb56b9
Merge pull request #15658 from BerriAI/litellm_response_mode_add_health
...
Add support for responses mode in health check
2025-10-17 09:44:34 -07:00
Ishaan Jaffer
09a2e32663
test_completion_cohere_command_r_plus_function_call
2025-10-17 09:40:14 -07:00
Ishaan Jaffer
80af02a1c3
test fix: tests.test_litellm.proxy.guardrails.guardrail_hooks.guardrails_ai.test_guardrails_ai::test_guardrails_ai_process_input
2025-10-17 09:39:32 -07:00
Sameer Kankute
282eedb72b
Add responses mode to health check
2025-10-17 22:06:29 +05:30
YutaSaito
bb3b77cbeb
feat: add guardrail for image generation ( #15619 )
...
* feat: add guardrail for image generation
* fix: use get_str_from_messages
2025-10-16 21:50:30 -07:00
Lucas Sugi
be8854f62e
fix: Convert object to a correct type ( #15634 )
2025-10-16 21:47:32 -07:00
Nicholas Couture
8032e73872
[Fix] Ensure guardrail memory sync after database updates ( #15633 )
...
* chore: Consistency in install-test-deps using poetry run
* feat: update in-memory guardrails after database CRUD operations
* test: add parameterized tests for guardrail CRUD with memory sync
2025-10-16 21:46:49 -07:00
Krrish Dholakia
814aeb5fad
fix(pass_through_endpoints.py): ensure passthrough endpoints are loaded in from db correctly on restarts
2025-10-16 21:39:48 -07:00
Krish Dholakia
3ff073c811
UI - add arize on ui, LLMs - clarifai refactor to openai compatible route, added azure ai/grok-4 model family
...
* added oauth mcp to docs
* added azure ai/grok-4 model family
* Revert "added oauth mcp to docs"
This reverts commit 950b7cef44d6c877b73a#15615 )
* fix: npm build errors
* Snowflake provider support: added embeddings, PAT, account_id (#15372 )
* snowflake support PAT, account_id and embeddings
* format
* test embeddings
* format
* complete test
---------
Co-authored-by: Krish Dholakia <krrishdholakia@gmail.com>
* Revert "Snowflake provider support: added embeddings, PAT, account_id (#15372 )" (#15632 )
This reverts commit c6d58e5b4a
2025-10-16 20:39:15 -07:00
Ishaan Jaffer
39b0e76a33
test_completion_cohere_command_r_plus_function_call
2025-10-16 20:33:41 -07:00
Ishaan Jaffer
64876634ef
ui new build
2025-10-16 20:32:51 -07:00
Ishaan Jaffer
7a1f654da6
ui new build
2025-10-16 18:37:42 -07:00
Ishaan Jaffer
13e895b9b0
test fix
2025-10-16 18:00:46 -07:00
Ishaan Jaffer
1c21b2e285
fixes: AI/ML API
2025-10-16 17:52:37 -07:00
Ishaan Jaffer
1c53a686d5
fix e2eui
2025-10-16 17:48:27 -07:00
Ishaan Jaffer
bf20ecab80
Revert "fix get_tags_from_request_body"
...
This reverts commit 1c044dad99
2025-10-16 17:40:08 -07:00
Ishaan Jaffer
96449b319c
fix get_tags_from_request_body
2025-10-16 17:40:08 -07:00
Ishaan Jaff
02758ebc28
Merge pull request #15553 from BerriAI/litellm_oct_staging2
...
[Staging] Oct Staging branch 2
2025-10-16 17:38:20 -07:00
Ishaan Jaffer
ea69f4547d
Merge branch 'main' into litellm_oct_staging2
2025-10-16 17:06:29 -07:00
Ishaan Jaffer
00e6ec4985
bump: version 1.78.2 → 1.78.3
2025-10-16 16:57:04 -07:00
Ishaan Jaff
f69f7d101b
Merge pull request #15618 from BerriAI/litellm_bedrock_invoke_support
...
[Feat] Allow calling /invoke, /converse routes through AI Gateway + models on config.yaml
2025-10-16 16:56:23 -07:00
Ishaan Jaffer
a31a8108b5
fix pass thu ruff check
2025-10-16 16:45:14 -07:00
Ishaan Jaffer
0bef295b15
doc fix
2025-10-16 16:36:24 -07:00
Ishaan Jaffer
bbf9d6a57b
docs boto3 instructions
2025-10-16 16:34:56 -07:00
Ishaan Jaffer
79516563a9
docs add /invoke and /converse routes
2025-10-16 16:25:27 -07:00
Ishaan Jaffer
8c5d6891d4
docs pt
2025-10-16 16:18:13 -07:00
Ishaan Jaffer
fe5dd23a84
docs boto3 pass thoughs
2025-10-16 16:17:02 -07:00
Ishaan Jaffer
c2b69e371c
docs pass through
2025-10-16 16:13:08 -07:00
Ishaan Jaffer
f87c86806c
docs invoke
2025-10-16 16:13:01 -07:00
Ishaan Jaff
5009a43b0e
Merge pull request #15608 from BerriAI/litellm_sso_add_pkce
...
[Feat] UI SSO - Add PKCE for OKTA SSO
2025-10-16 15:48:10 -07:00
Ishaan Jaff
f27f2d4803
Merge branch 'main' into litellm_sso_add_pkce
2025-10-16 15:48:01 -07:00
Ishaan Jaffer
bcf53c6ceb
working - errors from bedrock through pass throughs
2025-10-16 15:40:42 -07:00
Krrish Dholakia
e43c3b9b88
test: add unit testing
2025-10-16 15:37:26 -07:00
Ishaan Jaffer
51f1907e98
test_bedrock_error_handling_returns_actual_error
2025-10-16 15:36:51 -07:00
Krrish Dholakia
b85aeb08ce
fix(ui_sso.py): support both 'app_roles' and 'roles' claim on jwt decode for SSO handler
2025-10-16 15:29:34 -07:00
Ishaan Jaffer
e17636d4a6
add _extract_model_from_bedrock_endpoint
2025-10-16 15:26:32 -07:00
Ishaan Jaffer
cc5eac4965
fix _add_deployment_model_to_endpoint_for_llm_passthrough_route
2025-10-16 15:10:54 -07:00
Ishaan Jaff
f98f299854
Merge pull request #15617 from BerriAI/litellm_october_alexsander_stanging
...
[OCT] Alexsander PERF improvements
2025-10-16 15:10:02 -07:00
Ishaan Jaffer
e5956ff0d4
handle_bedrock_passthrough_router_model
2025-10-16 15:08:06 -07:00
Ishaan Jaffer
902cf69ffd
fixes for async pass throughs
2025-10-16 14:59:00 -07:00
AlexsanderHamir
bdca4db941
Merge branch 'litellm_october_alexsander_stanging' of https://github.com/BerriAI/litellm into litellm_october_alexsander_stanging
2025-10-16 14:43:30 -07:00
AlexsanderHamir
aad0efd99a
fix(proxy): change check_file_size_under_limit to accept Collection[str]
...
Fixes mypy type error where router_model_names (a set) was incompatible
with List[str] parameter type.
Changed check_file_size_under_limit to accept Collection[str] instead of
List[str] since the parameter is only used for membership checks, making
it more flexible and avoiding unnecessary type conversions.
Resolves: proxy/proxy_server.py:5107 arg-type error
2025-10-16 14:42:27 -07:00
AlexsanderHamir
220fa3feeb
Add missing env key
2025-10-16 14:42:27 -07:00
AlexsanderHamir
e154ab6a66
test: ensure model_names is a O(1) datastructure
2025-10-16 14:42:24 -07:00
AlexsanderHamir
3b14866d78
perf(router): optimize model lookups with O(1) data structures
...
Replace O(n) scans with index map lookups and convert model_names to set. Standardize timing calls.
2025-10-16 14:41:59 -07:00
AlexsanderHamir
76b272dffa
test(router): update error message assertion after string concat optimization
...
Update test_generate_model_id_with_deployment_model_name to accept the new
error message format that results from the list+join optimization.
The function still correctly rejects None values with a TypeError, but the
error message changed from 'unsupported operand type(s) for +=' to
'expected str instance, NoneType found' due to the implementation change
from string concatenation to list joining.
2025-10-16 14:41:59 -07:00
AlexsanderHamir
f842b2a6ce
perf(router): optimize string concatenation in hash generation
...
Replace string concatenation in loop with list append + join pattern.
This improves time complexity from O(n²) to O(n) and avoids creating
many temporary string objects during hash ID generation.
2025-10-16 14:41:59 -07:00
AlexsanderHamir
8c58d165a7
perf(router): use shallow copy instead of deepcopy for model aliases
...
Replace copy.deepcopy() with dict.copy() in _get_all_deployments when
creating model aliases.
Safe because:
1. Only modifies top-level 'model_name' field (isolated by shallow copy)
2. Nested dicts (litellm_params, model_info) are never modified after return
3. When model_alias=None, returns original dict with NO copy, proving
callers expect nested structures to be read-only
4. Key insight: Code that needs to modify nested structures does deepcopy FIRST.
This proves the contract is: \"treat returned deployments as read-only for
nested fields.\" (see line 6894: copy.deepcopy before modifying litellm_params)
Performance: ~10-100x faster than deepcopy on nested dict structures.
Tests: All 114 router unit tests pass.
2025-10-16 14:41:58 -07:00
AlexsanderHamir
35de05dc8b
test: add static analysis to prevent O(n) linear scans in router
...
Add AST-based test to detect 'for ... in self.model_list' anti-pattern.
Enforces use of index maps (model_id_to_deployment_index_map and
model_name_to_deployment_indices) for O(1) lookups instead of O(n) iteration.
2025-10-16 14:41:58 -07:00
