scriptoriumadmin
  • Joined on 2026-03-01
scriptoriumadmin synced commits to refs/pull/49377/head at scriptoriumadmin/litellm from mirror 2026-11-01 07:04:14 +00:00
6909f27886 fix(logging): scrub pydantic cut sides that hold the prompt's own quotes
e83a7bfe4b fix(logging): keep the request's redaction opt-in when another callback param is rejected
Compare 2 commits »
scriptoriumadmin synced commits to refs/pull/49371/head at scriptoriumadmin/litellm from mirror 2026-11-01 07:04:14 +00:00
49f479e311 test: make monthly budget reset coverage deterministic
2b9f03b712 fix(router): preserve screenshot image routing
dfb45ecfda fix(router): inspect image content without translating routed requests
1595adfb2f docs(model-catalog): explain meter, band, channel and When (#49447)
3cb2e60de0 fix(cost): split a provider-reported total between input and output by the cost map rate ratio (#49357)
Compare 34 commits »
scriptoriumadmin synced commits to refs/pull/49370/merge at scriptoriumadmin/litellm from mirror 2026-11-01 07:04:14 +00:00
1595adfb2f docs(model-catalog): explain meter, band, channel and When (#49447)
3cb2e60de0 fix(cost): split a provider-reported total between input and output by the cost map rate ratio (#49357)
b2647116ca chore(cost-map): add openai gpt-6-astra-law from the pricing page (#49444)
8ff495be96 fix(proxy): price OpenAI pass-through /v1/decisions calls like native Decisions (#49392)
Compare 26 commits »
scriptoriumadmin synced commits to refs/pull/49367/merge at scriptoriumadmin/litellm from mirror 2026-11-01 07:04:14 +00:00
3129e44848 ci: cut the weekly rc branch on Monday 3am PT instead of Friday (#49472)
d610dee7d0 docs(pr-template): add a handwritten 1-3 sentence description section (#49475)
ced995df6e fix(logging): drop a first-token time left over from a previous attempt (#47541)
3155aacda4 fix(caching): stop counting an unrun server tool call mark, and follow #49192 and #35602 in their integration tests (#49468)
Compare 51 commits »
scriptoriumadmin synced commits to refs/pull/49364/merge at scriptoriumadmin/litellm from mirror 2026-11-01 07:04:14 +00:00
8b9c9632f0 fix(router): fall back through plain fallbacks when context_window_fallbacks does not name the group
87e32e7c6d fix(agents): show non-admin callers the proxy URL in agent cards (#49483)
9312a25f18 feat(chat): add deep research with persistent citations (#49508)
36dbb228d9 fix(streaming): preserve served models in OpenRouter streams (#49498)
Compare 63 commits »
scriptoriumadmin synced commits to refs/pull/49364/head at scriptoriumadmin/litellm from mirror 2026-11-01 07:04:14 +00:00
8b9c9632f0 fix(router): fall back through plain fallbacks when context_window_fallbacks does not name the group
fc4ad87a6c fix(router): read the context window fallbacks through the typed fallback list helper
04a0f9ae6f Merge remote-tracking branch 'origin/main' into litellm_bedrock_mantle_gpt_oss_context_overflow
c43a1addf2 refactor(router): extract PythonCore behind RouterCore and pass the router its config (#49119)
a5f980600c fix(a2a): answer proxy refusals on /a2a POST routes as JSON-RPC errors (#49476)
Compare 54 commits »
scriptoriumadmin synced commits to refs/pull/49361/merge at scriptoriumadmin/litellm from mirror 2026-11-01 07:04:14 +00:00
0b71cc8e0c fix(model_prices): correct perplexity agent api pricing and add gemini-4-argon, nemotron-3.5-super
1a9c933615 Merge remote-tracking branch 'origin/main' into litellm-providers/price-sync-databricks
7e4403ea66 fix(a2a): send agent litellm_params.api_key as bearer auth on the A2A hop (#49474)
6f86b117b5 fix(moderations): stop the router image check from 500ing on a list input (#49454)
Compare 51 commits »
scriptoriumadmin synced commits to refs/pull/49381/merge at scriptoriumadmin/litellm from mirror 2026-11-01 07:04:14 +00:00
e111c17c1e fix(logging): apply a callback's own message redaction to its failure events
74b9bc5a40 fix(proxy): keep the request's redaction opt-in on proxy-only failure logs
16c57fa6f5 fix(otel): keep only the request text per span so a failed request's span can be freed
0e9a8ff83d fix(proxy): scrub the prompt out of proxy failure hook errors when message logging is redacted
Compare 9 commits »
scriptoriumadmin synced commits to refs/pull/49381/head at scriptoriumadmin/litellm from mirror 2026-11-01 07:04:14 +00:00
74b9bc5a40 fix(proxy): keep the request's redaction opt-in on proxy-only failure logs
e111c17c1e fix(logging): apply a callback's own message redaction to its failure events
16c57fa6f5 fix(otel): keep only the request text per span so a failed request's span can be freed
0e9a8ff83d fix(proxy): scrub the prompt out of proxy failure hook errors when message logging is redacted
ae41f691a0 test(mlflow): fake only the MLflow client in the failure event test
Compare 8 commits »
scriptoriumadmin synced commits to refs/pull/49360/merge at scriptoriumadmin/litellm from mirror 2026-11-01 07:04:14 +00:00
a1b32c00ea Merge remote-tracking branch 'origin/main' into litellm_usage_export_key_user_metadata
187f936cb3 fix(proxy)!: support reload with multiple workers (#45909)
a1c5829f6b fix(bedrock): raise GovCloud West Llama 3 max output to the model card 8K (#49543)
f4f5e2c80f fix(cost-map): add 2026-11-04 deprecation_date to together_ai Qwen3.8-2.4T-A95B (#49542)
Compare 86 commits »
scriptoriumadmin synced commits to refs/pull/49360/head at scriptoriumadmin/litellm from mirror 2026-11-01 07:04:13 +00:00
a1b32c00ea Merge remote-tracking branch 'origin/main' into litellm_usage_export_key_user_metadata
187f936cb3 fix(proxy)!: support reload with multiple workers (#45909)
a1c5829f6b fix(bedrock): raise GovCloud West Llama 3 max output to the model card 8K (#49543)
f4f5e2c80f fix(cost-map): add 2026-11-04 deprecation_date to together_ai Qwen3.8-2.4T-A95B (#49542)
68dbf54b09 chore(e2e): remove the unused arize_phoenix callback from the stage mirror config (#49538)
Compare 85 commits »
scriptoriumadmin synced commits to refs/pull/49338/head at scriptoriumadmin/litellm from mirror 2026-11-01 07:04:13 +00:00
5a32061c71 test(sdk): terminate the HTTP stream fixture explicitly
deb0ceee99 fix(responses): keep affinity item IDs consistent across stream events
7e4403ea66 fix(a2a): send agent litellm_params.api_key as bearer auth on the A2A hop (#49474)
6f86b117b5 fix(moderations): stop the router image check from 500ing on a list input (#49454)
941f52f26f fix(proxy): preserve discovery after budget exhaustion (#48809) (#49463)
Compare 63 commits »
scriptoriumadmin synced commits to refs/pull/49337/merge at scriptoriumadmin/litellm from mirror 2026-11-01 07:04:13 +00:00
d8db1a41a1 refactor(proxy): define the test_model conversations in Python instead of JSON files
ddab201d68 test(translation): expect the EOF error frame when an Anthropic json_schema stream is cut (#49521)
1d270a31b7 refactor(proxy): keep the test_model system prompt in system_prompt.txt so the request files stay readable
22d0b1e443 fix(router): restore four base behaviors the RoutingContext refactor changed (#49395)
Compare 74 commits »
scriptoriumadmin synced commits to refs/pull/49337/head at scriptoriumadmin/litellm from mirror 2026-11-01 07:04:13 +00:00
d8db1a41a1 refactor(proxy): define the test_model conversations in Python instead of JSON files
1d270a31b7 refactor(proxy): keep the test_model system prompt in system_prompt.txt so the request files stay readable
ee4cc1c3be feat(proxy): fail test_model prompt caching when turn 3 reads fewer cached tokens than turn 2
cadd100677 feat(proxy): give the test_model messages fixture the Claude Code cache_control and system reminder shape
9d57ff0376 feat(proxy): send a mid-conversation system reminder on test_model turn 3
Compare 11 commits »
scriptoriumadmin synced commits to refs/pull/49324/merge at scriptoriumadmin/litellm from mirror 2026-11-01 07:04:13 +00:00
159a8e201f fix(proxy): leave out guardrail response headers past 8 KB on pass-through routes
6a71d578c3 Merge branch 'litellm_lakera_v2_gemini_contents' into litellm_passthrough_applied_guardrails_header
3084da3bc4 Merge branch 'litellm_passthrough_auth_guardrails' into litellm_lakera_v2_gemini_contents
8ec5c7ddf2 perf(proxy): record applied guardrails in time linear in the names a request carries
Compare 98 commits »
scriptoriumadmin synced commits to refs/pull/49324/head at scriptoriumadmin/litellm from mirror 2026-11-01 07:04:13 +00:00
159a8e201f fix(proxy): leave out guardrail response headers past 8 KB on pass-through routes
6a71d578c3 Merge branch 'litellm_lakera_v2_gemini_contents' into litellm_passthrough_applied_guardrails_header
3084da3bc4 Merge branch 'litellm_passthrough_auth_guardrails' into litellm_lakera_v2_gemini_contents
8ec5c7ddf2 perf(proxy): record applied guardrails in time linear in the names a request carries
44f3a6eafc fix(guardrails): read a pass-through body with chat messages or input as chat in the Lakera v2 guardrail
Compare 97 commits »
scriptoriumadmin synced commits to refs/pull/49323/merge at scriptoriumadmin/litellm from mirror 2026-11-01 07:04:13 +00:00
3084da3bc4 Merge branch 'litellm_passthrough_auth_guardrails' into litellm_lakera_v2_gemini_contents
8ec5c7ddf2 perf(proxy): record applied guardrails in time linear in the names a request carries
44f3a6eafc fix(guardrails): read a pass-through body with chat messages or input as chat in the Lakera v2 guardrail
cfcec9bde5 Merge branch 'litellm_passthrough_auth_guardrails' into litellm_lakera_v2_gemini_contents
Compare 80 commits »
scriptoriumadmin synced commits to refs/pull/49323/head at scriptoriumadmin/litellm from mirror 2026-11-01 07:04:13 +00:00
3084da3bc4 Merge branch 'litellm_passthrough_auth_guardrails' into litellm_lakera_v2_gemini_contents
8ec5c7ddf2 perf(proxy): record applied guardrails in time linear in the names a request carries
44f3a6eafc fix(guardrails): read a pass-through body with chat messages or input as chat in the Lakera v2 guardrail
cfcec9bde5 Merge branch 'litellm_passthrough_auth_guardrails' into litellm_lakera_v2_gemini_contents
377124588e fix(proxy): keep the guardrails a client names in metadata on guarded pass-through keys and merge guardrail lists in linear time
Compare 79 commits »