stanfordnlp/dspy 3.1.1

on GitHub

What's Changed

RLMs

• feat(rlm): Add RLM Module and improve PythonInterpreter by @isaacbmiller in #9193
• Use language in system instructions for dspy.Code fields by @mariusarvinte in #9106
• Update dspy.RLM to improve reliability and avoid pydantic warnings by @okhat in #9210
• fix(RLM): Change FinalAnswerResult to FinalOutput and remove RLM call by @isaacbmiller in #9212

GEPA

• Fix GEPAFeedbackMetric Protocol missing 'self' parameter by @Copilot in #9111
• feat(gepa): add tool description optimization for multi-agent systems by @Ju-usc in #8928
• Update pyproject.toml to update gepa from 0.0.22 -> 0.0.24 by @LakshyAAAgrawal in #9161
• Patch dspy.gepa to handle list[dict] output from a dspy.LM by @mariusarvinte in #9169

Adapters

• Upgrade json_repair to fix parsing issue of underscore-style float number by @chenmoneygithub in #9088
• Properly yield the last chunk in streaming by @chenmoneygithub in #9089
• Avoid unwanted exception chaining when running async tool in sync context by @stevapple in #9092
• Enhance StreamListener to support generic type annotations for output by @TomeHirata in #9112
• FIX: streamify was appending StatusStreamingCallback directly to the shared settings.callbacks list by @glesperance in #9073
• fix(BAMLAdapter): Use docstrings to describe BaseModels by @BenMcH in #9125
• Fix Responses API structured outputs by @Olocool17 in #9130

Maintenance

• Fix UsageTracker AttributeError when using ParallelExecutor with dspy.context by @Copilot in #9095
• Update uv.lock dspy version to match the latest release by @chenmoneygithub in #9067
• [docs] Add Google-style docstrings for dspy/adapters/chat_adapter.py ChatAdapter class #9063 by @azai91 in #9072
• Fix ContextWindowExceededError after 3 retries in react loop by @Copilot in #9110
• fix: skip merging when both usage entry values are None by @chizukicn in #9121
• Fix: enforce positive memory_max_entries for in-memory cache by @mshr-h in #9128
• Add typing to dspy.datasets.dataset by @max-muoto in #9143
• Increase max_iters for dspy.ReAct by @chenmoneygithub in #9162
• fix(dspy): prevent argument injection in LocalProvider subprocess calls by @Copilot in #9160
• fix(dspy): populate InputField default values in Predict by @ritsuki1227 in #9167
• fix(PythonInterpreter): Remove overly permissive read permissions and add strict allow by @isaacbmiller in #9081
• Add save / load methods for DSpy settings by @WeichenXu123 in #9165

Docs

• [doc] Correct conversation history tutorial example by @zhiyanliu in #9071
• fix: Resolve broken hover-nlp/hover dataset by @vincentkoc in #9069
• Fix missing Real-World Examples and Experimental RL Optimization sections on tutorials page by @0xRaduan in #8982
• Refactor ReflectiveExample TypedDict documentation by @LakshyAAAgrawal in #9133
• [Automatic] Update api reference by @chenmoneygithub in #9135
• Replace ujson with orjson in docs by @togimoto in #9148

New Contributors

• @zhiyanliu made their first contribution in #9071
• @azai91 made their first contribution in #9072
• @0xRaduan made their first contribution in #8982
• @Ju-usc made their first contribution in #8928
• @mariusarvinte made their first contribution in #9106
• @Olocool17 made their first contribution in #9130
• @chizukicn made their first contribution in #9121
• @mshr-h made their first contribution in #9128
• @max-muoto made their first contribution in #9143
• @togimoto made their first contribution in #9148
• @ritsuki1227 made their first contribution in #9167
• @WeichenXu123 made their first contribution in #9165

Full Changelog: 3.1.0...3.1.1