v0.15.0

What's Changed

• Cleaning up GSM8k's CalculatorEnv a bit by @jamesbraza in #164
• Testing in check with Frame by @jamesbraza in #166
• Lock file maintenance by @renovate in #168
• Update all non-major dependencies by @renovate in #167
• Latest tooling and fixed some test warnings by @jamesbraza in #170
• Lock file maintenance by @renovate in #171
• Forbid extra arguments in configuration by @sidnarayanan in #173
• Pulling in Ruff 0.9 by @jamesbraza in #172
• Allowing open answer prompting with MultipleChoiceQuestion by @jamesbraza in #175

Full Changelog: v0.14.0...v0.15.0

v0.14.0

What's Changed

• Lock file maintenance by @renovate in #160
• Created MultipleChoiceQuestion and MultipleChoiceEvaluation by @jamesbraza in #157
• Moved to extract_answer from #148 and back to gpt-4o-mini by @jamesbraza in #161

Full Changelog: v0.13.0...v0.14.0

v0.13.0

What's Changed

• Import as written in readme for functional doesn't work by @SamCox822 in #150
• Support multiple images by @ludomitch in #151
• Lock file maintenance by @renovate in #146
• Lock file maintenance by @renovate in #153
• minor adjustment to image testing to support png by @ludomitch in #152
• Renovate lockFileMaintenance respecting schedule and 2-week stability period by @jamesbraza in #154
• eval_answer allowing for str | EvalAnswerMode not just EvalAnswerMode by @jamesbraza in #156
• Updated description of content by @maykcaldas in #158

New Contributors

• @SamCox822 made their first contribution in #150
• @ludomitch made their first contribution in #151

Full Changelog: v0.12.0...v0.13.0

v0.12.0

What's Changed

• Update astral-sh/setup-uv action to v4 by @renovate in #134
• Update all non-major dependencies by @renovate in #133
• Fixing httpx after #133 automerge by @jamesbraza in #137
• Lock file maintenance by @renovate in #136
• Fixed codeflash's formatter-cmds by @jamesbraza in #135
• Added complex types to str by @whitead in #140
• Tool exection timeout by @sidnarayanan in #141
• Moved eval_answer from tools to core by @maykcaldas in #142
• Type hinted maybe_wait_for return by @jamesbraza in #143
• Clean codeflash.yml config by @misrasaurabh1 in #144
• Populated LICENSE by @jamesbraza in #147

New Contributors

• @maykcaldas made their first contribution in #142
• @misrasaurabh1 made their first contribution in #144

Full Changelog: v0.11.0...v0.12.0

v0.11.0

What's Changed

• Implements new functional syntax by @whitead in #132

Here's a quick example - see README For complete description.

from aviary import fenv


@fenv.start()
def my_env(topic):
    # return first observation, and the starting environment state
    # (empty in this case)
    return f"Write a story about {topic}", {}

@my_env.tool()
def multiply(x: float, y: float) -> float:
    """Multiply two numbers."""
    return x * y


@my_env.tool()
def print_story(story: str | bytes, state) -> None:
    """Print a story to user and complete task."""
    print(story)
    state.reward = 1
    state.done = True

env = my_env(topic="foo")
obs, tools = await env.reset()

Full Changelog: v0.10.2...v0.11.0

v0.10.2

What's Changed

• mypy ignoring cloudpickle as it's untyped by @jamesbraza in #128
• Handling ordered deprecation by @sidnarayanan in #130
• Making uvicorn import lazy by @sidnarayanan in #131

Full Changelog: v0.10.1...v0.10.2

v0.10.1

What's Changed

• Refactor specialize_content to use guard clauses by @chgonz0 in #119
• Lock file maintenance by @renovate in #121
• Fixing sorting of not ordered tool calls by @jamesbraza in #122
• Lock file maintenance by @renovate in #124
• Opting into ruff preview rules and validate-pyproject-schema-store hook swap by @jamesbraza in #126
• Lock file maintenance by @renovate in #125
• Renamed Environment.exec_tool_calls ordered to concurrency by @jamesbraza in #123
• Generic TaskDataset server/client by @sidnarayanan in #127

New Contributors

• @chgonz0 made their first contribution in #119

Full Changelog: v0.10.0...v0.10.1

v0.10.0

What's Changed

• Unify answer tool naming by @albertbou92 in #104
• Lock file maintenance by @renovate in #108
• Lock file maintenance by @renovate in #111
• Lock file maintenance by @renovate in #112
• Python 3.13 support by @jamesbraza in #113
• Replacing list[Messages] with Messages in environments by @sidnarayanan in #114
• Moving to uv dependency groups by @jamesbraza in #109
• Removed unused SQLAlchemy dev dependency by @jamesbraza in #110
• Setting HotPotQA proxy from environment variable by @jamesbraza in #105
• HotPotQA better search tool failure message by @jamesbraza in #107
• Adding fetch-depth to actions/checkout to fix CI by @jamesbraza in #120
• Added invalid tool checker by @whitead in #115

Full Changelog: v0.9.1...v0.10.0

v0.9.1

What's Changed

• Fix eval mode setting in config by @sidnarayanan in #103

Full Changelog: v0.9.0...v0.9.1

v0.9.0

What's Changed

• Updated messages explanatory table by @jamesbraza in #94
• Aviary Environment Building Tutorial by @Ryan-Rhys in #91
• aviary.core.EnvStateMessage by @sidnarayanan in #95
• Lock file maintenance by @renovate in #97
• Moving from large-runner to ubuntu-latest for CI by @jamesbraza in #98
• HotPotQA lookup returns the wrong result is asked to look up the same term in 2 different pages sequentially by @albertbou92 in #99
• Add EvalAnswerMode to HotPotQAEnv by @sidnarayanan in #102

New Contributors

• @Ryan-Rhys made their first contribution in #91
• @albertbou92 made their first contribution in #99

Full Changelog: v0.8.3...v0.9.0