agentjido / req_llm / bb96ca81c6c3cc6cf7186753c033b5a9f4d23e38-PR-220

Builds Branch Commit Type Ran Committer Via Coverage
bb96ca81c6c3cc6cf7186753c033b5a9f4d23e38-PR-220 dependabot/hex/jsv-0.11.5 deps(deps): bump jsv from 0.11.4 to 0.11.5 Bumps [jsv](https://github.com/lud/jsv) from 0.11.4 to 0.11.5. - [Changelog](https://github.com/lud/jsv/blob/main/CHANGELOG.md) - [Commits](https://github.com/lud/jsv/compare/v0.11.4...v0.11.5) --- upda... Pull #220 web-flow github
50.13
aa2682570247c381c9c18b57f5416737347fe28e main Fix reasoning_overlay pattern match for llmdb structure (#219) Fix reasoning_overlay pattern match for llmdb structure The reasoning_overlay() test helper was pattern matching on `reasoning: true` but llmdb stores the capability as `reasoning: %... push web-flow github
50.13
633a089adae6ee8af90ba7e4232a8c0e0e804b86-PR-219 fix-reasoning-overlay-pattern-match Fix reasoning_overlay to match actual llmdb structure Change pattern match from reasoning: true to reasoning: %{enabled: true} to align with the actual llmdb capabilities structure. This fixes the issue where reasoning models were not having thei... Pull #219 neilberkman github
50.13
932bb71a22339354d13a7a831534226c0b6df5c6 main feat: Add credential fallback for fixture recording (#218) Implements automatic fallback to existing fixtures when credentials are missing during RECORD mode for providers requiring cloud credentials. - Add optional `credential_missing?/1` callb... push web-flow github
49.96
cf6e1b8f7b1881dcd99c730caa00091cd5d8a995-PR-218 feat/fixture-credential-fallback-rebased feat: Add credential fallback for fixture recording Implements automatic fallback to existing fixtures when credentials are missing during RECORD mode for providers requiring cloud credentials. - Add optional `credential_missing?/1` callback to ... Pull #218 neilberkman github
49.96
367e0ffad7f2cfcfa2b07ab36a5797ac93acf43e-PR-218 feat/fixture-credential-fallback-rebased feat: Add credential fallback for fixture recording Implements automatic fallback to existing fixtures when credentials are missing during RECORD mode for providers requiring cloud credentials. - Add optional `credential_missing?/1` callback to ... Pull #218 neilberkman github
49.92
a47489b2d5819649e80a12521c32c25cd6afa60b main Fix cached token extraction from Google API responses (#192) Google's API returns cachedContentTokenCount in the usageMetadata field for both implicit (automatic) and explicit (CachedContent API) prompt caching. The provider was not extracting th... push web-flow github
50.01
53f23896b3b425fbc5e22a27c506c086cd5c5cec-PR-192 fix/google-cached-tokens Fix test to use ReqLLM.model/1 after LLMDB integration Pull #192 neilberkman github
50.01
4908869ad3982fb484b75fd9cbfb45cb808438d4 main Add Google Vertex AI Anthropic provider support (#217) Enables Claude models (Haiku 4.5, Sonnet 4.5, Opus 4.1) on Google Vertex AI through provider alias system. Single google_vertex module now serves both google_vertex and google_vertex_anthropi... push web-flow github
49.98
c0113d7fa9acf80c97cbecda1306b64683b9b274-PR-217 record-bedrock-vertex-fixtures Update FakeRegistry to return LLMDB.Model structs FakeRegistry was returning strings but LLMDB returns model structs, causing a type mismatch after the LLMDB integration. Updated FakeRegistry to return minimal LLMDB.Model structs with id and prov... Pull #217 neilberkman github
49.98
28395890dddea1cc1ed122b36a3aedc99632d7a1-PR-217 record-bedrock-vertex-fixtures Update FakeRegistry to return LLMDB.Model structs FakeRegistry was returning strings but LLMDB returns model structs, causing a type mismatch after the LLMDB integration. Updated FakeRegistry to return minimal LLMDB.Model structs with id and prov... Pull #217 neilberkman github
49.98
6795646508c4eeddacd6cccaa78d3bc995f20e64 main feat: Integrate llm_db for model metadata (v1.1.0) (#212) * Research * chore(req_llm): add llm_db dependency - Add llm_db path dependency with runtime: false - Upgrade zoi from 0.7 to 0.8 for llm_db compatibility - Add llm_db config to config.e... push web-flow github
50.3
74599224c8ccdd0b2269757cdee4e72470f2dff4-PR-212 breaking/reqllm-llmdb-forced Enhance Anthropic provider with additional options and update model handling - Introduced new provider schema for Anthropic, adding options such as `anthropic_top_k`, `stop_sequences`, and `thinking`. - Updated the `fetch_api_key_and_extra_option... Pull #212 mikehostetler github pending completion   set done
d201f2ef7a828fa76dd49cffe90035e4956f1d85 main fix anthropic extra options missing api_key (#216) push web-flow github
52.8
ea528ac24a72b79dd78ed89d67fc5b9a8c30e5a3-PR-216 main fix anthropic extra options missing api_key Pull #216 phcurado github
52.76