ldai_optimizer: v0.2.0

0.2.0 (2026-07-17)

⚠ BREAKING CHANGES

• Bump minimum LangChain version to 1.0.0

Features

• ability to specify variation key on config or from_options for optimization package (c5e39eb)
• ability to specify variation key on config or from_options for optimization package (#162) (f0c4612)
• add auto-commit option (4cb8859)
• add optimization for duration (5d76276)
• Add optimization package stub (#109) (ebd5166)
• add shared dataclass for calls so they can be handled by same handler (31c8385)
• add token limit handling (e8c6692)
• adds ability to optimize for cost (94de596)
• adds ability to optimize for cost (#172) (3b4baa3)
• adds ability to use inverted judges (0ed243e)
• adds ability to use inverted judges (#168) (23baeb4)
• Adds optimization package stub (58b7731)
• Adds optimization package stub (cc85a05)
• adds reporting for cost and latency optimization failures (365fa94)
• adds reporting for cost and latency optimization failures (#180) (d267832)
• all logs -> debug (2fd55e2)
• Bump minimum LangChain version to 1.0.0 (dc592c5)
• Drop support for python 3.9 (#114) (dc592c5)
• dx improvements for optimization package (7074cfa)
• ground truth optimization path (44c8c59)
• implement ability to use completions or agents for judge calls (ea596a7)
• implement additional config optimization api fields (#196) (4ed30d8)
• implement latency & token tracking for optimizations (288336e)
• Implement optimization code paths and functionality for initial release (#140) (5204c47)
• implementation of agent optimization + tests (1712e4f)
• implements LD API client, optimize_from_config path (2fecd54)
• implements optimize method in SDK, code moved (9859d08)
• partially implement optimize_from_config (d3e1f96)
• prevent overfitting via prompt changes and post-processing (8f9f1e2)
• use judge_passed for all calcs (a8f14de)

Bug Fixes

• address cursor feedback (f2f0894)
• adjust iteration logic so validation doesn't consume them (3042984)
• better handling of params and custom params for optimization (572a2aa)
• better handling of params and custom params for optimization (#163) (92f51fa)
• consistency with other makefiles (b9a5601)
• don't only evaluate final input in GT results (53f455f)
• don't only evaluate final input in GT results (9bedf9e)
• ensure cost data is persisted (4eb0bb0)
• fix for single-iteration optimizations (96eadb7)
• fix for single-iteration optimizations (#209) (ba5aa63)
• lint (aee6aa7)
• lint + missed variable rename (f8e5509)
• lints + structured output tool rename (8481690)
• move failure forward (3c65a6e)
• only strip known provider prefixes (66bc1f0)
• optimization: address Bugbot comments on PR #140 (9e57d86)
• optimization: fix CI test failures and apply Bugbot-reported fixes (ffad0cb)
• optimization: match model config by key as well as id for pricing lookup (a8415be)
• optimization: remove dead persistence counter code (c76cb16)
• optimization: retry LLM calls on transient provider errors (429/503/529) (a131e4b)
• optimization: use run-chosen context for config judges; guard non-dict JSON (64e12f2)
• pull model configs if available in options path (4fc1ecf)
• remove unnecessary token path (dc82818)
• sort imports (c032aaf)
• success path + add test, cursor feedback (8f3468f)

launchdarkly-server-sdk-ai: v1.2.0

1.2.0 (2026-07-17)

Features

• implements template/raw methods for fetching configs (1c06d35)
• implements template/raw methods for fetching configs (#204) (5106866)

launchdarkly-server-sdk-ai: v1.1.0

1.1.0 (2026-06-05)

Features

• Add region property to ModelConfig (45707b6)

launchdarkly-server-sdk-ai-openai: v0.7.0

0.7.0 (2026-05-19)

⚠ BREAKING CHANGES

• Remove async from create_model/agent/agent_graph methods (#187)

Features

• Remove async from create_model/agent/agent_graph methods (#187) (dddc00a)

launchdarkly-server-sdk-ai-langchain: v0.8.0

0.8.0 (2026-05-19)

⚠ BREAKING CHANGES

• Remove async from create_model/agent/agent_graph methods (#187)

Features

• Remove async from create_model/agent/agent_graph methods (#187) (dddc00a)

launchdarkly-server-sdk-ai: v1.0.1

1.0.1 (2026-05-19)

Bug Fixes

• Remove pre-release caution note from server-ai README (#194) (32d52c4)

launchdarkly-server-sdk-ai: v1.0.0

1.0.0 (2026-05-19)

⚠ BREAKING CHANGES

• Remove bedrock-specific tracker method, use track_metrics_of method with custom extractor (#191)
• Remove async from create_model/agent/agent_graph methods (#187)

Features

• Remove async from create_model/agent/agent_graph methods (#187) (dddc00a)
• Remove bedrock-specific tracker method, use track_metrics_of method with custom extractor (#191) (33e4a65)

launchdarkly-server-sdk-ai-openai: v0.6.1

0.6.1 (2026-05-14)

Bug Fixes

• Make judge runners non-multi-turn (#185) (5c21bd0)

launchdarkly-server-sdk-ai-openai: v0.6.0

0.6.0 (2026-05-13)

⚠ BREAKING CHANGES

• Narrow AgentGraphRunner.run input from Any to str (#177)
• Rename LDAIMetrics.usage and AIGraphMetrics.usage to .tokens (#175)
• rename GraphMetrics/GraphMetricSummary to AIGraphMetrics/AIGraphMetricSummary (#173)

Features

• Narrow AgentGraphRunner.run input from Any to str (#177) (cc7a0fe)
• rename GraphMetrics/GraphMetricSummary to AIGraphMetrics/AIGraphMetricSummary (#173) (583939d)
• Rename LDAIMetrics.usage and AIGraphMetrics.usage to .tokens (#175) (d8c4a70)

Bug Fixes

• Always return ModelSettings from _build_model_settings (#169) (6d1ce4c)

launchdarkly-server-sdk-ai-langchain: v0.7.1

0.7.1 (2026-05-14)

Bug Fixes

• Make judge runners non-multi-turn (#185) (5c21bd0)