Repository navigation
chore(sync): bring main to BerriAI main b41e6c966a - #8
Merged
Merged
Conversation
…rriAI#42592) * chore(prices): sync OpenRouter prices: 19 models, 9 new [18 held] openrouter/~deepseek/deepseek-flash-latest: input_cost_per_token, output_cost_per_token, cache_read_input_token_cost openrouter/~deepseek/deepseek-pro-latest: max_tokens, max_output_tokens, off_peak_pricing, input_cost_per_token, output_cost_per_token, cache_read_input_token_cost openrouter/~deepseek/deepseek-v4-flash-latest: input_cost_per_token, output_cost_per_token, cache_read_input_token_cost openrouter/~moonshotai/kimi-latest: input_cost_per_token, output_cost_per_token openrouter/~z-ai/glm-flash-latest: input_cost_per_token, output_cost_per_token, cache_read_input_token_cost openrouter/aion-labs/aion-2.0: max_input_tokens openrouter/aion-labs/aion-3.0: max_input_tokens openrouter/aion-labs/aion-3.0-mini: max_input_tokens openrouter/anthropic/claude-opus-5.5:batch: max_tokens, supports_vision, max_input_tokens, max_output_tokens, supports_pdf_input, supports_reasoning, supports_web_search, supports_audio_input, supports_tool_choice, supports_prompt_caching, supports_response_schema, supports_function_calling, input_cost_per_token, output_cost_per_token, cache_read_input_token_cost, cache_creation_input_token_cost, cache_creation_input_token_cost_above_1hr openrouter/cohere/command-a-plus: max_tokens, supports_vision, max_input_tokens, max_output_tokens, supports_pdf_input, supports_reasoning, supports_web_search, supports_audio_input, supports_tool_choice, supports_prompt_caching, supports_response_schema, supports_function_calling, input_cost_per_token, output_cost_per_token, cache_read_input_token_cost openrouter/deepseek/deepseek-v4-pro: input_cost_per_token, output_cost_per_token, cache_read_input_token_cost openrouter/deepseek/deepseek-v4.1-flash: max_tokens, max_output_tokens, off_peak_pricing openrouter/deepseek/deepseek-v4.1-flash:batch: max_tokens, supports_vision, max_input_tokens, max_output_tokens, supports_pdf_input, supports_reasoning, supports_web_search, supports_audio_input, supports_tool_choice, supports_prompt_caching, supports_response_schema, supports_function_calling, input_cost_per_token, output_cost_per_token, cache_read_input_token_cost openrouter/openai/gpt-6-luna-pro:batch: max_tokens, supports_vision, max_input_tokens, max_output_tokens, supports_pdf_input, supports_reasoning, supports_web_search, supports_audio_input, supports_tool_choice, supports_prompt_caching, supports_response_schema, supports_function_calling, input_cost_per_token, output_cost_per_token, cache_read_input_token_cost, cache_creation_input_token_cost, input_cost_per_token_above_272k_tokens, output_cost_per_token_above_272k_tokens, cache_read_input_token_cost_above_272k_tokens, cache_creation_input_token_cost_above_272k_tokens openrouter/openai/gpt-6-luna:batch: max_tokens, supports_vision, max_input_tokens, max_output_tokens, supports_pdf_input, supports_reasoning, supports_web_search, supports_audio_input, supports_tool_choice, supports_prompt_caching, supports_response_schema, supports_function_calling, input_cost_per_token, output_cost_per_token, cache_read_input_token_cost, cache_creation_input_token_cost, input_cost_per_token_above_272k_tokens, output_cost_per_token_above_272k_tokens, cache_read_input_token_cost_above_272k_tokens, cache_creation_input_token_cost_above_272k_tokens openrouter/openai/gpt-6-sol-pro:batch: max_tokens, supports_vision, max_input_tokens, max_output_tokens, supports_pdf_input, supports_reasoning, supports_web_search, supports_audio_input, supports_tool_choice, supports_prompt_caching, supports_response_schema, supports_function_calling, input_cost_per_token, output_cost_per_token, cache_read_input_token_cost, cache_creation_input_token_cost, input_cost_per_token_above_272k_tokens, output_cost_per_token_above_272k_tokens, cache_read_input_token_cost_above_272k_tokens, cache_creation_input_token_cost_above_272k_tokens openrouter/openai/gpt-6-sol:batch: max_tokens, supports_vision, max_input_tokens, max_output_tokens, supports_pdf_input, supports_reasoning, supports_web_search, supports_audio_input, supports_tool_choice, supports_prompt_caching, supports_response_schema, supports_function_calling, input_cost_per_token, output_cost_per_token, cache_read_input_token_cost, cache_creation_input_token_cost, input_cost_per_token_above_272k_tokens, output_cost_per_token_above_272k_tokens, cache_read_input_token_cost_above_272k_tokens, cache_creation_input_token_cost_above_272k_tokens openrouter/openai/gpt-oss-20b:batch: max_tokens, supports_vision, max_input_tokens, max_output_tokens, supports_pdf_input, supports_reasoning, supports_web_search, supports_audio_input, supports_tool_choice, supports_prompt_caching, supports_response_schema, supports_function_calling, input_cost_per_token, output_cost_per_token openrouter/qwen/qwen3.8-omni-flash: max_tokens, supports_vision, max_input_tokens, max_output_tokens, supports_pdf_input, supports_reasoning, supports_web_search, supports_audio_input, supports_tool_choice, supports_prompt_caching, supports_response_schema, supports_function_calling, input_cost_per_token, output_cost_per_token, cache_read_input_token_cost * chore(prices): sync OpenRouter prices: 1 model [9 held] openrouter/deepseek/deepseek-v4.1-flash: off_peak_pricing Price-Sync: litellm-providers * chore(prices): sync OpenRouter prices: 1 model [12 held] openrouter/deepseek/deepseek-v4.1-flash: off_peak_pricing Price-Sync: litellm-providers * chore(prices): sync OpenRouter prices: 1 model [9 held] openrouter/deepseek/deepseek-v4.1-flash: off_peak_pricing Price-Sync: litellm-providers * chore(prices): sync OpenRouter prices: 1 model [12 held] openrouter/deepseek/deepseek-v4.1-flash: off_peak_pricing Price-Sync: litellm-providers * chore(prices): sync OpenRouter prices: 1 model [10 held] openrouter/deepseek/deepseek-v4.1-flash: off_peak_pricing Price-Sync: litellm-providers * chore(prices): sync OpenRouter prices: 1 model [13 held] openrouter/deepseek/deepseek-v4-pro-0813: off_peak_pricing Price-Sync: litellm-providers * feat(prices): add openrouter/upstage/solar-mini4 from OpenRouter models API Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * chore(prices): sync OpenRouter prices: 1 model [16 held] openrouter/deepseek/deepseek-v4.1-flash: off_peak_pricing Price-Sync: litellm-providers * feat(prices): add openrouter/aion-labs/aion-3.5 Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> --------- Co-authored-by: berriai-litellm-provider-info-sync[bot] <328147090+berriai-litellm-provider-info-sync[bot]@users.noreply.github.com> Co-authored-by: kerry <kerry@berri.ai> Co-authored-by: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
… deprecation dates (BerriAI#42738)
…iAI#42701) PATCH /model/{id}/update rejected read-modify-write edits that resent an unchanged but dangling litellm_credential_name. Existence validation now runs only when the requested name differs from the stored one; empty string and non-admin detach rejections are unchanged Co-authored-by: yuneng <yuneng@berri.ai> Co-authored-by: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
…#42743) Co-authored-by: kerry <kerry@berri.ai> Co-authored-by: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
…cp shard and proxy coverage artifact (BerriAI#42711) * test(integration): drop the contracts.json manifest and the covers requirement Groups live as a GROUPS literal in run.py, the browser expectations move next to the browser tests, and the runner fails only on pytest failure, collection errors or a selected file that collects zero tests. The covers marker stays registered for the existing tests but is no longer checked. The mcp directory gets its own group Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * ci(integration): run mcp as its own shard with xdist and a peer proxy Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * ci(integration): INTEGRATION_COVERAGE=1 runs the proxy under coverage for the MCP modules The mcp shard sets it. The proxy and its peer start under coverage run in parallel mode, get SIGTERM after the tests so coverage flushes, and the combined text and HTML reports land in the suite results that CircleCI already stores as artifacts Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * ci(integration): let the test proxy flush coverage when uvicorn re-raises SIGTERM Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * test(integration): add SSE, stdio, scripted, OpenAPI and OAuth 2.1 MCP peer doubles Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * test(integration): add MCP transport and access-control matrices Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * test(integration): add MCP credential and OAuth flow coverage Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * test(integration): add MCP LLM endpoint, accounting, guardrail, resilience and lifecycle coverage Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * test(integration): stop the same-URL grant test from counting a late initialize as a leaked call and satisfy the test-tree lint Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * test(integration): assert the REST denied-server listing is refused or empty Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * test(integration): pin the REST denied-server listing to 403 access_denied Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> --------- Co-authored-by: yuneng <yuneng@berri.ai> Co-authored-by: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
…er casing on save (BerriAI#42703) Co-authored-by: yuneng <yuneng@berri.ai> Co-authored-by: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
…rences, main is the only trunk (BerriAI#42745)
BerriAI#42747) * feat(cost-map): add Azure Foundry pricing for gpt-6-sol and gpt-6-luna Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * fix(cost-map): give azure/eu gpt-6-sol and gpt-6-luna full model metadata Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> --------- Co-authored-by: kerry <kerry@berri.ai> Co-authored-by: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
…erriAI#42619) * fix(secrets): verify provider API request and payload contracts * wip * fix(secrets): unify backend reads and route secret resolution * feat(secrets): bind built-in managers to retained Rust backends * refactor(secrets): centralize catalog dispatch and native binding * test(secrets): split provider integration tests * refactor(secrets): enforce cache and rotation contracts * test(secrets): stub parent packages in failing resolver fixture Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * refactor(secrets): pass manager settings through the interop boundary Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * fix(secrets): align cloud KMS auth and harden provider reads * ci(rust): raise native wheel size gate to 40 MB for secrets backends Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * fix(proxy): treat unset google kms flag as disabled like the old loader Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * fix(secrets): preserve certificate credentials and disabled KMS flags * test(secrets): cover certificate validation and bounded auth retries * test(secrets): cover Python dispatch without the native extension * test(proxy): skip legacy secret manager cases when the optional SDK is missing Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * fix(secrets): port Python parity tests and preserve provider behavior * fix(secrets): store the captured native config without setattr to satisfy the strict lint budget Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * fix(secrets): preserve missing Azure manager values Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * test(secrets): pin typed values and recovery failure precedence * refactor(secrets): organize provider internals and behavioral test suites * refactor(secrets): simplify recovery and isolate Python compatibility * fix(secrets): distinguish Azure callback absence from HTTP not found * fix(secrets): preserve Python AWS read results at the bridge * fix(secrets): route public reads through the native catalog bridge * fix(secrets): keep JSON selection outside the bridge Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * fix(secrets): preserve provider JSON reads at the bridge Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * fix(secrets): preserve Python primary JSON semantics Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * fix(secrets): preserve CyberArk mutation behavior through the native bridge * docs(secrets): record public API replacement gaps * refactor(secrets): share Vault write payload preparation * feat(secrets): route Vault mutations through the native bridge * fix(secrets): preserve typed Vault rotation failures * refactor(secrets): move Python dispatch into bridge * refactor(secrets): move CyberArk Python policy into bridge * refactor(secrets): move Vault Python policy into bridge * test(secrets): assert Vault rotation request paths * fix(secrets): keep bridge JSON interop centralized Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> --------- Co-authored-by: Yujong Lee <yujong@berri.ai> Co-authored-by: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
…eout tests (rolling deflake 2026-09-22) (BerriAI#42125) * test(autoroute): wait for a valid fuzzy selection index and cancel the prompt on driver failure Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * test(autoroute): read the fuzzy selection through the public InquirerPy property Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * test(proxy): inject the HIBP client into change_password so the breached-password test never touches the network Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * test(mcp): only use the 200ms read timeout in the silent mode of the transport completion test Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * test(proxy): record HIBP requests so the ordering test asserts no lookup happened Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * ci(codeql): filter the weak-sensitive-data-hashing false positive on the HIBP k-anonymity lookup Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> --------- Co-authored-by: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
…prices (BerriAI#42752) * feat(gemini): add gemini-3.8-flash-tts and gemini-3.8-flash-lite-tts prices Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * fix(gemini): bill tiered TTS output through output_cost_per_token tiers Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> --------- Co-authored-by: kerry <kerry@berri.ai> Co-authored-by: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
…uting and streaming bugs (BerriAI#42693) * test(integration): optional Anthropic tool properties stay optional on the OpenAI Responses wire (Pylon BerriAI#6619) Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * test(integration): Bedrock InvokeModel count-suffixed cache usage fields are reported and charged (Pylon BerriAI#6708) Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * test(integration): anthropic messages honors the deployment request timeout (Pylon BerriAI#6505) Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * test(integration): drop client_metadata before the Bedrock Converse body reaches the provider (Pylon BerriAI#6645) Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * test(integration): repeat Bedrock requests under one session name assume the role once (Pylon BerriAI#6681) Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * test(integration): messages stream keeps include_usage off the Responses wire with always_include_stream_usage (Pylon BerriAI#6466) Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * test(integration): clamp sub-16 max_tokens to the Responses API floor instead of 400 (Pylon BerriAI#6539) Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * test(integration): forwarded client x- headers reach the provider on /v1/responses (Pylon BerriAI#6565) Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * test(integration): Anthropic messages stop_sequences reach OpenAI-compatible providers as stop (Pylon BerriAI#6536) Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * test(integration): Codex namespace tools reach a chat upstream flattened and round-trip through /v1/responses (Pylon BerriAI#6409) Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * test(integration): keep Claude 4.6 legacy thinking budget_tokens on /v1/messages (Pylon BerriAI#6727) Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * test(integration): nvidia nim ranking keeps image passages and applies top_n without sending top_k (Pylon BerriAI#6401) Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * test(integration): tpm-only model rejects priority traffic once recorded tokens reach the model tpm (Pylon BerriAI#6344) Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * test(integration): reasoning-only chunks open an Anthropic thinking block at index zero on /v1/messages streams (Pylon BerriAI#6337) Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * test(integration): file content streams to the client before the upstream finishes sending (Pylon BerriAI#6315) Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * test(integration): agent whose card lives only at agentCard/v1.0 is reached with bearer auth (Pylon BerriAI#6249) Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * test(integration): vertex batch create returns a batch when outputInfo is null (Pylon BerriAI#6374) Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * test(integration): fireworks session id is sent as x-session-affinity and cached tokens land in spend log metadata (Pylon BerriAI#6220) Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * test(integration): advisor sub-call failure does not cool down the executor deployment (Pylon BerriAI#6212) Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * test(integration): Gemini /v1/messages cache_control creates cachedContent with Anthropic ttl (Pylon BerriAI#6221) Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * test(integration): bedrock_mantle max_output_tokens below 16 is clamped before reaching Mantle (Pylon BerriAI#6262) Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * test(integration): missing thinking signature 400 on /v1/messages retries without thinking blocks (Pylon BerriAI#6222) Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * test(integration): format Gemini messages cache_control wire test (Pylon BerriAI#6221) Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * test(integration): rebuilt shared aiohttp session keeps the configured keepalive timeout (Pylon BerriAI#6387) Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * test(integration): sagemaker_chat signs the inference component header and sends hf_model_name as the body model (Pylon BerriAI#6187) Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * test(integration): Bedrock Converse DeepSeek drops Anthropic thinking and sends V3 reasoning_effort raw (Pylon BerriAI#6149) Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * test(integration): concurrent team model TPM requests are reserved before the provider call (Pylon BerriAI#6075) Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * test(integration): /v1/messages honors the configured timeout against a stalled upstream (Pylon BerriAI#6025) Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * test(integration): Codex additional_tools input items reach Bedrock Mantle as top-level tools (Pylon BerriAI#6012) Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * test(integration): advisor api_base without api_key never sends the proxy Anthropic key to the caller host (Pylon BerriAI#6226) Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * test(integration): rerank responses carry call id, latency and cost headers (Pylon BerriAI#5981) Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * test(integration): parse the outbound Anthropic body with the typed JSON adapter (Pylon BerriAI#6025) Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * test(integration): Bedrock Knowledge Base search forwards userContext to the Retrieve body (Pylon BerriAI#5991) Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * test(integration): Marengo 3.0 text embeddings reach Bedrock nested under inputType (Pylon BerriAI#5949) Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * test(integration): sub-16 max_tokens over a responses deployment reaches OpenAI as 16 (Pylon BerriAI#6008) Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * test(integration): midturn system correction reaches the OpenAI Responses wire via /v1/messages (Pylon BerriAI#6449) Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * test(integration): concurrent requests over a key tpm limit are rejected before reaching the provider (Pylon BerriAI#5737) Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * test(integration): chat to responses bridge keeps deployment AWS credentials for Bedrock Mantle SigV4 (Pylon BerriAI#5870) Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * test(integration): vertex gemini stream split across many fragments completes without stalling the proxy (Pylon BerriAI#5838) Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * test(integration): bedrock mantle /v1/messages stream keeps stream true and relays SSE events (Pylon BerriAI#5596) Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * test(integration): large chat payloads are released from worker memory after the request ends (Pylon BerriAI#5920) Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * test(integration): streaming success logs v3 rate limit remaining values for callbacks (Pylon BerriAI#5767) Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * test(integration): a database created search tool backs Anthropic web search interception (Pylon BerriAI#5669) Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * test(integration): register july provider regression contracts Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * test(integration): make july provider regression tests deterministic under cache and worker sharing Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * test(integration): drop order-fragile worker memory probe pending a real retention regression check Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * test(integration): apply ruff import sorting and formatting Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * test(integration): drop stale contract entry and pass question to advisor executor Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * test(integration): use tiktoken-backed executor model in advisor tests Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> --------- Co-authored-by: kerry <kerry@berri.ai> Co-authored-by: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
… and spend bugs (BerriAI#42694) * test(integration): streamed Bedrock Messages usage cost equals the recorded spend (Pylon BerriAI#6667) Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * test(integration): echoed cost-map model info is not persisted as deployment overrides (Pylon BerriAI#6844) Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * test(integration): reset sweep runs on one pod per tick while replicas share the lease (Pylon BerriAI#6521) Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * test(integration): bedrock post-call guardrail scans streamed Anthropic Messages tool use without 500 (Pylon BerriAI#6503) Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * test(integration): realtime cached audio tokens bill at the audio cache-read rate (Pylon BerriAI#6704) Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * test(integration): legacy GET /spend/logs returns at most the 10000 most recent rows and flags truncation (Pylon BerriAI#6752) Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * test(integration): itemize Responses API cache write tokens as cache creation cost (Pylon BerriAI#6454) Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * test(integration): migration entrypoint deploys pending migrations before proxy startup (Pylon BerriAI#6649) Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * test(integration): opted-in team keys stop at the owner's personal budget (Pylon BerriAI#6641) Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * test(integration): guardrail information stays in the spend log when the caller sends metadata on /v1/messages (Pylon BerriAI#6614) Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * test(integration): key model allowlist is enforced on Bedrock passthrough routes (Pylon BerriAI#6419) Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * test(integration): JWT mapped key backfills a null user email from token claims (Pylon BerriAI#6266) Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * test(integration): scheduled budget reset recovers from a transient DB transport failure (Pylon BerriAI#6582) Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * test(integration): team key lists models granted through a team access group (Pylon BerriAI#6044) Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * test(integration): JWT subject without team claim lands in the configured default team (Pylon BerriAI#5895) Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * test(integration): end-user spend lands for a key without user_id when the auth cache is Redis (Pylon BerriAI#6021) Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * test(integration): stale-low redis counter still blocks team member over budget (Pylon BerriAI#5824) Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * test(integration): prompt-carrying spend rows are written in byte-bounded statements (Pylon BerriAI#6083) Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * test(integration): logs UI session_total_spend sums every round of a multi-round session (Pylon BerriAI#5928) Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * test(integration): config.yaml guardrails are served by the guardrail usage detail and overview (Pylon BerriAI#5813) Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * test(integration): plain chat request skips the object permission lookup (Pylon BerriAI#5965) Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * test(integration): register july accounting regression contracts Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * test(integration): isolate cost map override clear on owned proxy Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * test(integration): make reset lease claim and db relay refusal deterministic Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * test(integration): poll pg_stat settle, bound unbanned relay refusals, clear reset lease on teardown Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * test(integration): bound relay refusals so the budget sweep can reconnect Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> --------- Co-authored-by: kerry <kerry@berri.ai> Co-authored-by: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
Co-authored-by: kerry <kerry@berri.ai> Co-authored-by: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
…erriAI#42664) * fix(params): carry stream_chunk_size through litellm_params instead of provider params * test(integration): fence stream_chunk_size out of every provider request body * test(bedrock): type parametrized stream chunk test params * test(integration): drop the contracts manifest resurrected by the main merge * test(bedrock): type the stream_chunk_size test helpers * test(params): finish AGENTS.md typing pass on stream_chunk_size tests * test(integration): drop the covers marker from the stream_chunk_size wire test --------- Co-authored-by: shrey kharbanda <shreshth@berri.ai>
Brings the branch to BerriAI/litellm main b41e6c9 so the PR is measured against the current base, and straiker-ai/litellm main stays even with BerriAI. No Straiker file changes. Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
PhimmStraiker
had a problem deploying
to
e2e-changed
September 23, 2026 17:29 — with 
GitHub Actions
Failure
GitHub Actions
Failure
Keeps straiker-ai/litellm main even with BerriAI. No Straiker file changes. Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
PhimmStraiker
had a problem deploying
to
e2e-changed
September 23, 2026 17:51 — with
GitHub Actions
Failure
GitHub Actions
Failure
rekiarts
approved these changes
Sep 23, 2026
This branch had an error being deployed
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Keeps
straiker-ai/litellmmaineven with BerriAI and identical to the head of BerriAI#41880.main@b41e6c966a(15 upstream commits since feat(straiker): v3 platform guardrail on current BerriAI main #7: model prices, integration tests, CI cleanup).mainand the BerriAI PR head are the same commit again (5378c7ece2).🤖 Generated with Claude Code