--- name: smolagents slug: smolagents type: github source_url: https://github.com/huggingface/smolagents organization: Hugging Face organization_slug: hugging-face total_releases: 34 latest_version: v1.24.0 latest_date: 2026-01-16 last_updated: 2026-04-19 tracking_since: 2024-12-27 canonical: https://releases.sh/hugging-face/smolagents organization_url: https://releases.sh/hugging-face --- Smolagents maintained compatibility across evolving dependencies while expanding model support. The release added backward compatibility for the deprecated HfApiModel, updated the no-stop-sequence model list to cover GPT-5.2 variants, and patched Gradio 6 compatibility in the UI layer. Support for additional parameters in chat template application rounded out the update. ## What's Changed * Bump version to 1.24 by @aymeric-roucher in https://github.com/huggingface/smolagents/pull/1871 * Fix: Add backward compatibility for deprecated HfApiModel by @MohammadKassas143 in https://github.com/huggingface/smolagents/pull/1863 * Fix typo in secure_code_execution.md by @jonathanagustin in https://github.com/huggingface/smolagents/pull/1892 * Update no-stop-sequence model list to support gpt-5.2* by @aymeric-roucher in https://github.com/huggingface/smolagents/pull/1895 * Fix run_gaia.py token_counts when managed agent is called more than once by @suryabdev in https://github.com/huggingface/smolagents/pull/1878 * ๐ŸŒ [i18n-KO] Translated `tools.md` to Korean by @Kim-Ju-won in https://github.com/huggingface/smolagents/pull/1835 * Support passing additional params to apply_chat_template by @albertvillanova in https://github.com/huggingface/smolagents/pull/1902 * fix(gradio_ui): Gradio 6 compatibility by @alejandro-ao in https://github.com/huggingface/smolagents/pull/1903 * Coerce tool calls from APIs into smolagents ChatMessageToolCall format by @aymeric-roucher in https://github.com/huggingface/smolagents/pull/1900 * Add FinalAnswerStep to possible step_callbacks by @suryabdev in https://github.com/huggingface/smolagents/pull/1881 * Upgrade GitHub Actions for Node 24 compatibility by @salmanmkc in https://github.com/huggingface/smolagents/pull/1909 * fix: prevent FinalAnswerException from being caught by except Exception by @majiayu000 in https://github.com/huggingface/smolagents/pull/1914 * Add DeepWiki badge to README by @aymeric-roucher in https://github.com/huggingface/smolagents/pull/1846 * Implement robust timeout mechanism for Python execution by @albertvillanova in https://github.com/huggingface/smolagents/pull/1910 ## New Contributors * @MohammadKassas143 made their first contribution in https://github.com/huggingface/smolagents/pull/1863 * @jonathanagustin made their first contribution in https://github.com/huggingface/smolagents/pull/1892 * @alejandro-ao made their first contribution in https://github.com/huggingface/smolagents/pull/1903 * @salmanmkc made their first contribution in https://github.com/huggingface/smolagents/pull/1909 * @majiayu000 made their first contribution in https://github.com/huggingface/smolagents/pull/1914 **Full Changelog**: https://github.com/huggingface/smolagents/compare/v1.23.0...v1.24.0 ## Improvements โœจ * Update final answer checks to accept the agent instance by @Zoe14 in https://github.com/huggingface/smolagents/pull/1648 * Dialog mode CLI by @aymeric-roucher in https://github.com/huggingface/smolagents/pull/1540 * Add Blaxel support for remote code execution by @Joffref in https://github.com/huggingface/smolagents/pull/1791 * Proposition: Add exponential backoff with jitter for retries by @suryabdev in https://github.com/huggingface/smolagents/pull/1829 * Cut generation at stop sequence post generation by @aymeric-roucher in https://github.com/huggingface/smolagents/pull/1802 * Move executor type validation to python executor creation by @Lrakotoson in https://github.com/huggingface/smolagents/pull/1800 * Support custom Python code executor in CodeAgent by @albertvillanova in https://github.com/huggingface/smolagents/pull/1707 * Retry on rate limit errors by @aymeric-roucher in https://github.com/huggingface/smolagents/pull/1801 * Change default InferenceClient model to Qwen/Qwen3-Next-80B-A3B-Thinking by @suryabdev in https://github.com/huggingface/smolagents/pull/1813 * Remove 'Server' from model names in doc by @aymeric-roucher in https://github.com/huggingface/smolagents/pull/1812 * Optimize comprehension evaluation with generator-based approach in LocalPythonExecutor by @albertvillanova in https://github.com/huggingface/smolagents/pull/1824 * LocalPythonExecutor: Add support for nested dictcomp and setcomp by @suryabdev in https://github.com/huggingface/smolagents/pull/1823 * Support parsing `anyOf` from MCP tools by @duguyue100 in https://github.com/huggingface/smolagents/pull/1708 * Add Modal as sandbox option to docs front page by @sandcat100 in https://github.com/huggingface/smolagents/pull/1865 * make additional_args nullable for managed agents by @nflo in https://github.com/huggingface/smolagents/pull/1827 * vLLM: Move from guided_options_request to structured_outputs by @suryabdev in https://github.com/huggingface/smolagents/pull/1805 ## Fixes ๐Ÿ› ๏ธ * Fix agent logs: logging "None" and simplify the logic by @albertvillanova in https://github.com/huggingface/smolagents/pull/1788 * Fix dict(message) bug in AgentMemory.replay for ChatMessage objects by @Flakes342 in https://github.com/huggingface/smolagents/pull/1763 * Fix LocalPythonExecutor support for Enum and other metaclasses by @colesmcintosh in https://github.com/huggingface/smolagents/pull/1797 * Handle `None` content in stop-sequence trimming by @chahn in https://github.com/huggingface/smolagents/pull/1826 * Fix minor benchmark script bugs by @suryabdev in https://github.com/huggingface/smolagents/pull/1822 * Fix YAML scanner error when building documentation by @albertvillanova in https://github.com/huggingface/smolagents/pull/1841 * Pin huggingface-hub <1.0.0 to fix failed to build tokenizers by @albertvillanova in https://github.com/huggingface/smolagents/pull/1844 * Fixed SyntaxError format: keep exception type and message on same line (#835) by @balazsthomay in https://github.com/huggingface/smolagents/pull/1864 * Fix CLI Tool.from_space() call by auto-generating name and description (#1535) by @balazsthomay in https://github.com/huggingface/smolagents/pull/1859 * Add gpt-5.1 support by @suryabdev in https://github.com/huggingface/smolagents/pull/1868 * bug: role not converted back to enum by @njbrake in https://github.com/huggingface/smolagents/pull/1833 * Parallel tool calls lose OpenTelemetry context: tool spans not nested under agent step by @Barcavin in https://github.com/huggingface/smolagents/pull/1840 ## Documentation ๐Ÿ“š * Bump dev version: v1.23.0.dev0 by @albertvillanova in https://github.com/huggingface/smolagents/pull/1787 * ๐ŸŒ [i18n-KO] Translated memory.md to Korean by @HyunZ118 in https://github.com/huggingface/smolagents/pull/1793 * ๐ŸŒ [i18n-KO] Translated `using_different_models.md` to Korean by @FacerAin in https://github.com/huggingface/smolagents/pull/1772 * ๐ŸŒ [i18n-KO] Translated `guided_tour.md` to Korean by @FacerAin in https://github.com/huggingface/smolagents/pull/1711 * ๐ŸŒ [i18n-KO] Translated `async_agent.md` to Korean by @FacerAin in https://github.com/huggingface/smolagents/pull/1749 * ๐ŸŒ [i18n-KO] Translated `agents.md` to Korean by @Kim-Ju-won in https://github.com/huggingface/smolagents/pull/1720 * ๐ŸŒ [i18n-KO] Translated `rag.md` to Korean by @Kim-Ju-won in https://github.com/huggingface/smolagents/pull/1646 * Documentation: Minor fixes by @suryabdev in https://github.com/huggingface/smolagents/pull/1809 * ๐ŸŒ [i18n-KO] Translated `web_browser.md` to Korean by @FacerAin in https://github.com/huggingface/smolagents/pull/1748 * [i18n-es] Translating docs to spanish by @1-echo in https://github.com/huggingface/smolagents/pull/1806 * ๐ŸŒ [i18n-KO] Translated `plan_customization.md` to Korean by @FacerAin in https://github.com/huggingface/smolagents/pull/1771 * ๐ŸŒ [i18n-KO] Translated `inspect_runs.md` to Korean by @FacerAin in https://github.com/huggingface/smolagents/pull/1747 * Build PR documentation for all languages by @albertvillanova in https://github.com/huggingface/smolagents/pull/1842 **Full Changelog**: https://github.com/huggingface/smolagents/compare/v1.22.0...v1.23.0 ## What's Changed * Bump dev version: v1.22.0.dev0 by @albertvillanova in https://github.com/huggingface/smolagents/pull/1657 * refactor(PythonExecutor): inherit from ABC and declare abstract methods by @HairlessVillager in https://github.com/huggingface/smolagents/pull/1654 * Fix error for GPT-5 not supporting the `stop` parameter by @Xmader in https://github.com/huggingface/smolagents/pull/1662 * Enable return_full_result directly in run method by @aymeric-roucher in https://github.com/huggingface/smolagents/pull/1658 * Make RunResult json serializable by @aymeric-roucher in https://github.com/huggingface/smolagents/pull/1659 * Change truncation to remove end rather than middle by @aymeric-roucher in https://github.com/huggingface/smolagents/pull/1665 * Document that model kwargs are forwarded to model completion call by @albertvillanova in https://github.com/huggingface/smolagents/pull/1679 * Fix missing kwargs forward to vLLM model generate call by @albertvillanova in https://github.com/huggingface/smolagents/pull/1680 * Raise RuntimeError with LiteLLM response details if no choices by @albertvillanova in https://github.com/huggingface/smolagents/pull/1683 * Fix AmazonBedrockModel with reasoning/thinking content by @albertvillanova in https://github.com/huggingface/smolagents/pull/1681 * Fix installation command in README by quoting extra by @shahgahmed in https://github.com/huggingface/smolagents/pull/1685 * CI hotfix: Pin openai < 1.100.0 for litellm extra by @albertvillanova in https://github.com/huggingface/smolagents/pull/1693 * Support openai 1.100 for litellm extra by @albertvillanova in https://github.com/huggingface/smolagents/pull/1698 * Pin telemetry extra >=0.1.15 to use new TokenUsage structure by @albertvillanova in https://github.com/huggingface/smolagents/pull/1699 * Add AGENTS.md by @albertvillanova in https://github.com/huggingface/smolagents/pull/1701 * Test telemetry extra by @albertvillanova in https://github.com/huggingface/smolagents/pull/1700 * Replace slim with bullseye base image in Docker executor by @albertvillanova in https://github.com/huggingface/smolagents/pull/1702 * fix: error missing f-string by @njbrake in https://github.com/huggingface/smolagents/pull/1691 * Support custom Dockerfile in DockerExecutor by @albertvillanova in https://github.com/huggingface/smolagents/pull/1705 * Support MCP structured output and output schema by @chahn in https://github.com/huggingface/smolagents/pull/1663 * Improve agent image handling by @aaron-ang in https://github.com/huggingface/smolagents/pull/1601 * ๐ŸŒ [i18n-KO] Translated `multiagents.md` to Korean by @pyapyapya in https://github.com/huggingface/smolagents/pull/1644 * Fix CI 403 error for Wikipedia page in test_visit_webpage by @albertvillanova in https://github.com/huggingface/smolagents/pull/1716 * fix typos in the system prompt by @Frame17 in https://github.com/huggingface/smolagents/pull/1710 * Fix pip install commands by quoting extras by @albertvillanova in https://github.com/huggingface/smolagents/pull/1690 * Fix deprecation warning about parameter messages by @albertvillanova in https://github.com/huggingface/smolagents/pull/1718 * Fix warning about parameter max_new_tokens by @albertvillanova in https://github.com/huggingface/smolagents/pull/1721 * Adds Modal Remote Executor by @thomasjpfan in https://github.com/huggingface/smolagents/pull/1722 * Fix formatting of title in translation section of documentation by @albertvillanova in https://github.com/huggingface/smolagents/pull/1727 * Remove untranslated sections from the documentation by @albertvillanova in https://github.com/huggingface/smolagents/pull/1729 * fix: add ipykernel in dockerfile of prebuilt docker executor by @cyyeh in https://github.com/huggingface/smolagents/pull/1731 * Fix E2BExecutor by pinning e2b-code-interpreter to < 2 by @albertvillanova in https://github.com/huggingface/smolagents/pull/1733 * Fix DockerExecutor connection reset error with server readiness check by @benoriol in https://github.com/huggingface/smolagents/pull/1684 * [i18n-KO] Translated building_good_agents.md to Korean by @FacerAin in https://github.com/huggingface/smolagents/pull/1714 * ๐ŸŒ [i18n-KO] Translated installation.md to Korean by @FacerAin in https://github.com/huggingface/smolagents/pull/1647 * Support e2b-code-interpreter v2 by @August-murr in https://github.com/huggingface/smolagents/pull/1751 * Fix token count initialization when plan_message.token_usage is None by @satwikkansal in https://github.com/huggingface/smolagents/pull/1752 * Support user-configurable parameter override for model completion parameters by @albertvillanova in https://github.com/huggingface/smolagents/pull/1678 * Fix error for grok-4/grok-3-mini not supporting the stop parameter by @suryabdev in https://github.com/huggingface/smolagents/pull/1766 * Replace DockerExecutor Dockerfile with file object by @albertvillanova in https://github.com/huggingface/smolagents/pull/1738 * Update documentation to clarify `instructions` are appended by @chahn in https://github.com/huggingface/smolagents/pull/1709 * ๐ŸŒ [i18n-KO] Translated `react.md` to Korean by @nsbg in https://github.com/huggingface/smolagents/pull/1666 * Pass tools_to_call_from to TransformersModel._prepare_completion_kwargs by @albertvillanova in https://github.com/huggingface/smolagents/pull/1782 * ๐ŸŒ [i18n-KO] Translated `models.md` to Korean by @Kim-Ju-won in https://github.com/huggingface/smolagents/pull/1776 * Fix AttributeError when trying to log a None by @albertvillanova in https://github.com/huggingface/smolagents/pull/1786 * fix(agent): remove unused get_variable_names function by @keli-wen in https://github.com/huggingface/smolagents/pull/1784 * Fix XPath injection in search_item_ctrl_f by @albertvillanova in https://github.com/huggingface/smolagents/pull/1768 * Use context manager for per-operation websocket connection in DockerExecutor by @albertvillanova in https://github.com/huggingface/smolagents/pull/1750 ## New Contributors * @Xmader made their first contribution in https://github.com/huggingface/smolagents/pull/1662 * @shahgahmed made their first contribution in https://github.com/huggingface/smolagents/pull/1685 * @aaron-ang made their first contribution in https://github.com/huggingface/smolagents/pull/1601 * @pyapyapya made their first contribution in https://github.com/huggingface/smolagents/pull/1644 * @Frame17 made their first contribution in https://github.com/huggingface/smolagents/pull/1710 * @cyyeh made their first contribution in https://github.com/huggingface/smolagents/pull/1731 * @satwikkansal made their first contribution in https://github.com/huggingface/smolagents/pull/1752 * @suryabdev made their first contribution in https://github.com/huggingface/smolagents/pull/1766 * @Kim-Ju-won made their first contribution in https://github.com/huggingface/smolagents/pull/1776 * @keli-wen made their first contribution in https://github.com/huggingface/smolagents/pull/1784 **Full Changelog**: https://github.com/huggingface/smolagents/compare/v1.21.3...v1.22.0 ## Bug Fixes ๐Ÿ› * Fix E2BExecutor by pinning e2b-code-interpreter to < 2 by @albertvillanova in https://github.com/huggingface/smolagents/pull/1733 **Full Changelog**: https://github.com/huggingface/smolagents/compare/v1.21.2...v1.21.3 ## Bug Fixes ๐Ÿ› * Pin telemetry extra >=0.1.15 to use new TokenUsage structure by @albertvillanova in https://github.com/huggingface/smolagents/pull/1699 **Full Changelog**: https://github.com/huggingface/smolagents/compare/v1.21.1...v1.21.2 ## Bug Fixes ๐Ÿ› * Fix error for GPT-5 not supporting the `stop` parameter by @Xmader in https://github.com/huggingface/smolagents/pull/1662 **Full Changelog**: https://github.com/huggingface/smolagents/compare/v1.21.0...v1.21.1 ## Enhancements โœจ * **Tool & Model Improvements** * Add Tool to_code_prompt and to_tool_calling_prompt by @albertvillanova in https://github.com/huggingface/smolagents/pull/1599 * Improve function body extraction in tools by @HairlessVillager in https://github.com/huggingface/smolagents/pull/1627 * Support passing model_kwargs to TransformersModel by @albertvillanova in https://github.com/huggingface/smolagents/pull/1608 * Support dict messages as direct input to models by @albertvillanova in https://github.com/huggingface/smolagents/pull/1570 * **Executor Improvements** * Enhance `LocalPythonExecutor` security by blocking dunder calls by @nnfrog in https://github.com/huggingface/smolagents/pull/1551 * Fix `continue` semantics of `LocalPythonExecutor`by @adaminsky in https://github.com/huggingface/smolagents/pull/1645 * Fix `GeneratorExp` in `LocalPythonExecutor` to return `Generator` instead of `list` by @albertvillanova in https://github.com/huggingface/smolagents/pull/1650 * **Gradio & UI** * Reset agent memory on GradioUI clear button by @albertvillanova in https://github.com/huggingface/smolagents/pull/1611 * Use `gr.Number` for mapping integer and number type components in `launch_gradio_demo` by @karenwky in https://github.com/huggingface/smolagents/pull/1320 ## Bug Fixes ๐Ÿ› * Fix: Prevent empty additional_args from being added to agent prompt by @jokokojote in https://github.com/huggingface/smolagents/pull/1566 * Fix SyntaxError in generated Gradio app due to 'class' attribute by @albertvillanova in https://github.com/huggingface/smolagents/pull/1615 * Fix serialization inconsistency in MultiStepAgent.to_dict/from_dict for managed_agents by @hazrmard in https://github.com/huggingface/smolagents/pull/1577 * Fix a kwargs error in LocalPythonExecutor by @aymeric-roucher in https://github.com/huggingface/smolagents/pull/1583 * Fix structured outputs by @MicrosoftJames in https://github.com/huggingface/smolagents/pull/1642 * Fix bug where `AmazonBedrockServerModel` crashes in thinking mode by @benoriol in https://github.com/huggingface/smolagents/pull/1632 * Fix: Remove duplicate line in `tool_calling_agent` prompt by @yusuf-eren in https://github.com/huggingface/smolagents/pull/1636 * Fix: `RemotePythonExecutor` should not send variables if none exist by @thomasjpfan in https://github.com/huggingface/smolagents/pull/1605 * Disable parallel tool calls for final answer by @aymeric-roucher in https://github.com/huggingface/smolagents/pull/1539 ## Documentation ๐Ÿ“š * Improve Installation docs with uv option by @albertvillanova in https://github.com/huggingface/smolagents/pull/1550 * Fix and improve MCP server security warnings in docs by @albertvillanova in https://github.com/huggingface/smolagents/pull/1549 * DOC: Fix image link for docs (secure_code_execution.md) by @FacerAin in https://github.com/huggingface/smolagents/pull/1582 * Update title of page "What are agents" by @aymeric-roucher in https://github.com/huggingface/smolagents/pull/1546 * Enable multilingual docs by @albertvillanova in https://github.com/huggingface/smolagents/pull/1633 * ๐ŸŒ [i18n-KO] Translated `index.md` to Korean by @FacerAin in https://github.com/huggingface/smolagents/pull/1581 * ๐ŸŒ [i18n-KO] Translated `text_to_sql.md` to Korean by @nsbg in https://github.com/huggingface/smolagents/pull/1603 ## CI / Dependency Updates ๐Ÿงช * Pin mcpadapt to support image/audio content by @albertvillanova in https://github.com/huggingface/smolagents/pull/1558 * Temporarily skip wrongly implemented tests by @albertvillanova in https://github.com/huggingface/smolagents/pull/1571 * CI hotfix: Pin mlx < 0.26.5 by @albertvillanova in https://github.com/huggingface/smolagents/pull/1586 * Support mlx-0.26.5 by @albertvillanova in https://github.com/huggingface/smolagents/pull/1589 * CI hotfix: Pin transformers < 4.54.0 by @albertvillanova in https://github.com/huggingface/smolagents/pull/1620 * Support transformers 4.54.0 by @albertvillanova in https://github.com/huggingface/smolagents/pull/1625 * Fix CI PytestUnknownMarkWarning: Unknown pytest.mark.timeout by @albertvillanova in https://github.com/huggingface/smolagents/pull/1630 * Fix CI quality: remove trailing whitespace by @albertvillanova in https://github.com/huggingface/smolagents/pull/1617 ## Deprecations & Cleanups ๐Ÿงน * Replace deprecated ast.Num and ast.Str with ast.Constant in tool validation by @WilliamRabuel in https://github.com/huggingface/smolagents/pull/1594 * Remove deprecated `grammar` parameter by @albertvillanova in https://github.com/huggingface/smolagents/pull/1651 * Remove deprecated token count attributes by @albertvillanova in https://github.com/huggingface/smolagents/pull/1652 * Remove deprecated agent `logs` attribute by @albertvillanova in https://github.com/huggingface/smolagents/pull/1653 * Remove mention of deprecated `ManagedAgent` from docs by @albertvillanova in https://github.com/huggingface/smolagents/pull/1655 * Remove deprecated default `sse` transport by @albertvillanova in https://github.com/huggingface/smolagents/pull/1656 * Replace duckduckgo_search with ddgs to address package rename warning by @WilliamRabuel in https://github.com/huggingface/smolagents/pull/1593 * Remove unused is_same_item function by @albertvillanova in https://github.com/huggingface/smolagents/pull/1631 * Remove unnecessary import by @rahul-1729 in https://github.com/huggingface/smolagents/pull/1639 * Simplify _is_package_available by @albertvillanova in https://github.com/huggingface/smolagents/pull/1641 ## Miscellaneous ๐Ÿ”ง * Add support for Amazon Bedrock API Keys by @colesmcintosh in https://github.com/huggingface/smolagents/pull/1612 * Factorize create_agent_gradio_app_template function by @albertvillanova in https://github.com/huggingface/smolagents/pull/1616 * Fix: Allow passing integers for floats by @aymeric-roucher in https://github.com/huggingface/smolagents/pull/1623 * Fix Tool.forward to raise NotImplementedError instead of returning by @HairlessVillager in https://github.com/huggingface/smolagents/pull/1628 * Make validate_tool_arguments raise instead of returning error string by @albertvillanova in https://github.com/huggingface/smolagents/pull/1634 * Remove mention of python in additional args by @aymeric-roucher in https://github.com/huggingface/smolagents/pull/1584 * ToolCallingAgent: remove unnecessary addition to `model_output` by @aymeric-roucher in https://github.com/huggingface/smolagents/pull/1622 * Minor fix in toolCallingAgentPrompt by @rahul-1729 in https://github.com/huggingface/smolagents/pull/1574 ## Maintenance ๐Ÿ—๏ธ * Bump dev version: v1.21.0.dev0 by @albertvillanova in https://github.com/huggingface/smolagents/pull/1544 * Remove empty issue template by @albertvillanova in https://github.com/huggingface/smolagents/pull/1562 * Improve issue templates by @albertvillanova in https://github.com/huggingface/smolagents/pull/1563 * Add issue template for documentation improvements by @albertvillanova in https://github.com/huggingface/smolagents/pull/1564 * Add uv option to CONTRIBUTING by @albertvillanova in https://github.com/huggingface/smolagents/pull/1561 ## New Contributors * @jokokojote made their first contribution in https://github.com/huggingface/smolagents/pull/1566 * @rahul-1729 made their first contribution in https://github.com/huggingface/smolagents/pull/1574 * @WilliamRabuel made their first contribution in https://github.com/huggingface/smolagents/pull/1594 * @FacerAin made their first contribution in https://github.com/huggingface/smolagents/pull/1582 * @thomasjpfan made their first contribution in https://github.com/huggingface/smolagents/pull/1605 * @hazrmard made their first contribution in https://github.com/huggingface/smolagents/pull/1577 * @HairlessVillager made their first contribution in https://github.com/huggingface/smolagents/pull/1628 * @benoriol made their first contribution in https://github.com/huggingface/smolagents/pull/1632 * @yusuf-eren made their first contribution in https://github.com/huggingface/smolagents/pull/1636 * @nnfrog made their first contribution in https://github.com/huggingface/smolagents/pull/1551 * @nsbg made their first contribution in https://github.com/huggingface/smolagents/pull/1603 * @karenwky made their first contribution in https://github.com/huggingface/smolagents/pull/1320 * @MicrosoftJames made their first contribution in https://github.com/huggingface/smolagents/pull/1642 * @adaminsky made their first contribution in https://github.com/huggingface/smolagents/pull/1645 **Full Changelog**: https://github.com/huggingface/smolagents/compare/v1.20.0...v1.21.0 ## New Features โœจ * **Remote Execution**: * Implement remote Python WasmExecutor by @albertvillanova in https://github.com/huggingface/smolagents/pull/1261 * **Callbacks**: * Support callbacks after planning step by @albertvillanova in https://github.com/huggingface/smolagents/pull/1445 * Support multiple callbacks per step type in step_callbacks dict by @albertvillanova in https://github.com/huggingface/smolagents/pull/1517 ## Enhancements ๐Ÿ› ๏ธ * **Models**: * Implement rate limiting in all API models by @albertvillanova in https://github.com/huggingface/smolagents/pull/1516 * **Tools**: * Implement exception mechanism for final answer in remote executors by @tobiasofsn in https://github.com/huggingface/smolagents/pull/1429 * Set default api_key_name in ApiWebSearchTool by @albertvillanova in https://github.com/huggingface/smolagents/pull/1507 * Implement rate limit mechanism in ApiWebSearchTool by @albertvillanova in https://github.com/huggingface/smolagents/pull/1508 * Implement rate limit mechanism in DuckDuckGoSearchTool by @albertvillanova in https://github.com/huggingface/smolagents/pull/1515 * Enable image output for Tool.from_space by @aymeric-roucher in https://github.com/huggingface/smolagents/pull/1510 * Add adapter_kwargs parameter to MCPClient by @LucasDltg in https://github.com/huggingface/smolagents/pull/1528 * **Validation of Tool Arguments**: * Support multiple types in tool arguments validation by @duguyue100 in https://github.com/huggingface/smolagents/pull/1505 * Fix input type validation when input schema has multiple types by @duguyue100 in https://github.com/huggingface/smolagents/pull/1530 * Fix optional args toolcalling by @aymeric-roucher in https://github.com/huggingface/smolagents/pull/1538 * **Agent Output Formatting**: * Implement CodeOutput as analog to ToolOutput by @albertvillanova in https://github.com/huggingface/smolagents/pull/1496 * Allow markdown or custom formatting for code blocks by @aymeric-roucher in https://github.com/huggingface/smolagents/pull/1493 * **Other**: * Add ApiWebSearchTool to __all__ list by @albertvillanova in https://github.com/huggingface/smolagents/pull/1492 * Enforced type annotations in `ChatMessage` via `MessageRole` enum by @daavoo in https://github.com/huggingface/smolagents/pull/1479 * Check authorized imports in LocalPythonExecutor constructor by @jank in https://github.com/huggingface/smolagents/pull/265 ## Bug Fixes ๐Ÿ› * Fix `KeyError` when agent reaches max steps with image input by @robin-ede in https://github.com/huggingface/smolagents/pull/1525 * Fix when `response.usage` is `None` for some OpenRouter models by @sentienthouseplant in https://github.com/huggingface/smolagents/pull/1402 * Fix incorrect token counting in streaming TransformersModel by @albertvillanova in https://github.com/huggingface/smolagents/pull/1503 * Minor fix nested quotes that prevent execution by @vinoo-ganesh-keru in https://github.com/huggingface/smolagents/pull/1497 * Fix ChatMessage attribute access after replacing dict with dataclass by @albertvillanova in https://github.com/huggingface/smolagents/pull/1534 * Fix access of `content` field in `ChatMessage` by @mdr223 in https://github.com/huggingface/smolagents/pull/1533 * Fix parallel final answers handling by @aymeric-roucher in https://github.com/huggingface/smolagents/pull/1482 ## Documentation Improvements ๐Ÿ“š * **Examples & Tutorials**: * Add complete Plan Customization example by @albertvillanova in https://github.com/huggingface/smolagents/pull/1506 * Add Async App with Agent example by @albertvillanova in https://github.com/huggingface/smolagents/pull/1509 * Improve Plan Customization example with Human-in-the-Loop by @albertvillanova in https://github.com/huggingface/smolagents/pull/1513 * **API References**: * Fix docstring of WikipediaSearchTool by @albertvillanova in https://github.com/huggingface/smolagents/pull/1521 * Explain rate_limit in tools docstrings by @albertvillanova in https://github.com/huggingface/smolagents/pull/1519 * Fix Examples in AmazonBedrockServerModel docstring by @albertvillanova in https://github.com/huggingface/smolagents/pull/1522 * Split Built-in Tools from Tools in Reference docs by @albertvillanova in https://github.com/huggingface/smolagents/pull/1520 * **Other**: * Update Langfuse tracing documentation to SDK v3 by @jannikmaierhoefer in https://github.com/huggingface/smolagents/pull/1447 * Fix use of context manager agent in docs by @albertvillanova in https://github.com/huggingface/smolagents/pull/1502 ## Maintenance ๐Ÿ—๏ธ * Fix CI LiteLLM test_call_different_providers_without_key by @albertvillanova in https://github.com/huggingface/smolagents/pull/1527 * Fix test_send_tools_with_default_wikipedia_search_tool by @albertvillanova in https://github.com/huggingface/smolagents/pull/1494 * Fix DockerExecutor tests with final_answer by calling send_tools by @albertvillanova in https://github.com/huggingface/smolagents/pull/1495 * Bump dev version: v1.20.0.dev0 by @albertvillanova in https://github.com/huggingface/smolagents/pull/1477 ## New Contributors * @sentienthouseplant made their first contribution in https://github.com/huggingface/smolagents/pull/1402 * @daavoo made their first contribution in https://github.com/huggingface/smolagents/pull/1479 * @duguyue100 made their first contribution in https://github.com/huggingface/smolagents/pull/1505 * @vinoo-ganesh-keru made their first contribution in https://github.com/huggingface/smolagents/pull/1497 * @robin-ede made their first contribution in https://github.com/huggingface/smolagents/pull/1525 * @LucasDltg made their first contribution in https://github.com/huggingface/smolagents/pull/1528 * @mdr223 made their first contribution in https://github.com/huggingface/smolagents/pull/1533 **Full Changelog**: https://github.com/huggingface/smolagents/compare/v1.19.0...v1.20.0 ## Enhancements ๐Ÿ› ๏ธ * **Agent Upgrades**: * Support managed agents in ToolCallingAgent by @albertvillanova in https://github.com/huggingface/smolagents/pull/1456 * Support context managers for agent cleanup by @tobiasofsn in https://github.com/huggingface/smolagents/pull/1422 * Change code tags to xml by @aymeric-roucher in https://github.com/huggingface/smolagents/pull/1442 * **UI Improvements**: * Support reset_agent_memory in GradioUI by @JakeBx in https://github.com/huggingface/smolagents/pull/1420 * **Streaming Refactor**: * Transfer aggregation of streaming events off the Model class by @aymeric-roucher in https://github.com/huggingface/smolagents/pull/1449 * **Agent Output Tracking**: * Store CodeAgent code outputs in ActionStep by @aymeric-roucher in https://github.com/huggingface/smolagents/pull/1463 ## Bug Fixes ๐Ÿ› * Fix Agent update planning logic by @Zoe14 in https://github.com/huggingface/smolagents/pull/1417 * Remove plural from named argument return_full_results in examples by @vladlen32230 in https://github.com/huggingface/smolagents/pull/1434 * Fix and refactor final answer checks by @aymeric-roucher in https://github.com/huggingface/smolagents/pull/1448 * Fix logging of Docker build logs by @tobiasofsn in https://github.com/huggingface/smolagents/pull/1421 * Add a mention of additional_args in the manager agent's prompt by @aymeric-roucher in https://github.com/huggingface/smolagents/pull/1441 * Fix missing mentions of additional_args in manager agent prompts by @albertvillanova in https://github.com/huggingface/smolagents/pull/1459 * Fix: `__new__` method only accepts class object, remove **args and **kwโ€ฆ by @abdulhakkeempa in https://github.com/huggingface/smolagents/pull/1462 * Do not wrap types in safer_func by @albertvillanova in https://github.com/huggingface/smolagents/pull/1475 * Match multiline final answers in remote executors by @albertvillanova in https://github.com/huggingface/smolagents/pull/1444 * Revert removal of the last message from memory_messages during planning by @Zoe14 in https://github.com/huggingface/smolagents/pull/1454 ## Documentation Improvements ๐Ÿ“š * **Quickstart Enhancements**: * Add simple tool usage in doc quickstart by @aymeric-roucher in https://github.com/huggingface/smolagents/pull/1470 * Add open-colab link to Quickstart docs by @albertvillanova in https://github.com/huggingface/smolagents/pull/1472 * **Reference Docs**: * Add AgentMemory to Reference docs by @albertvillanova in https://github.com/huggingface/smolagents/pull/1473 * Add docstrings to GradioUI by @albertvillanova in https://github.com/huggingface/smolagents/pull/1451 * **Fixes & Formatting**: * Correct broken link in documentation homepage by @johntony366 in https://github.com/huggingface/smolagents/pull/1468 * Rename docs files from .mdx to .md by @albertvillanova in https://github.com/huggingface/smolagents/pull/1471 ## Maintenance ๐Ÿ—๏ธ * Bump dev version: v1.19.0.dev0 by @albertvillanova in https://github.com/huggingface/smolagents/pull/1427 ## New Contributors * @Zoe14 made their first contribution in https://github.com/huggingface/smolagents/pull/1417 * @vladlen32230 made their first contribution in https://github.com/huggingface/smolagents/pull/1434 * @JakeBx made their first contribution in https://github.com/huggingface/smolagents/pull/1420 * @abdulhakkeempa made their first contribution in https://github.com/huggingface/smolagents/pull/1462 * @johntony366 made their first contribution in https://github.com/huggingface/smolagents/pull/1468 **Full Changelog**: https://github.com/huggingface/smolagents/compare/v1.18.0...v1.19.0 ## New Features โœจ * **Multiple Parallel Tool Calls**: ToolCallingAgent can now handle multiple tool calls in parallel, significantly enhancing performance for complex tasks. * Support multiple tool calls in parallel in ToolCallingAgent by @albertvillanova in https://github.com/huggingface/smolagents/pull/1412 * **Streaming Output for ToolCallingAgent**: ToolCallingAgent now supports streaming outputs, improving responsiveness and user experience during multi-step tool interactions * Streaming outputs for ToolCallingAgent ๐Ÿš€ by @aymeric-roucher in https://github.com/huggingface/smolagents/pull/1409 * **API Web Search Tool**: Introduced a new ApiWebSearchTool, enabling structured web search capabilities via API. * Create API web search tool by @albertvillanova in https://github.com/huggingface/smolagents/pull/1400 ## Enhancements ๐Ÿ› ๏ธ * Update prompts to avoid confusion: search -> web_search and wiki -> wiki_search by @SrzStephen in https://github.com/huggingface/smolagents/pull/1403 * Make Agent.system_prompt read only by @albertvillanova in https://github.com/huggingface/smolagents/pull/1399 * Support configurable tool_choice in prepare_completion_kwargs by @albertvillanova in https://github.com/huggingface/smolagents/pull/1392 * Support passing additional params to MLXModel load and tokenizer.apply_chat_template by @albertvillanova in https://github.com/huggingface/smolagents/pull/1406 * Support custom headers/params for ApiWebSearchTool by @albertvillanova in https://github.com/huggingface/smolagents/pull/1411 ## Bug Fixes ๐Ÿ› * Fix: Support custom inputs execution for custom Final Answer Tool by @Lrakotoson in https://github.com/huggingface/smolagents/pull/1383 * Fix @tool decorator for remote Python executor by @tobiasofsn in https://github.com/huggingface/smolagents/pull/1334 * Always pass add_generation_prompt=True to apply_chat_template by @albertvillanova in https://github.com/huggingface/smolagents/pull/1416 ## Documentation Improvements ๐Ÿ“š * Make docs neutral about agent types by @julien-c in https://github.com/huggingface/smolagents/pull/1376 * Add MCP tools doc section by @aymeric-roucher in https://github.com/huggingface/smolagents/pull/1394 * Update ToolCollection.from_mcp docstring examples by @neonwatty in https://github.com/huggingface/smolagents/pull/1398 * Document better final_answer_checks by @albertvillanova in https://github.com/huggingface/smolagents/pull/1407 * Fix typo in secure code execution documentation by @chahn in https://github.com/huggingface/smolagents/pull/1414 * Add chat server demo to examples by @albertvillanova in https://github.com/huggingface/smolagents/pull/1415 * Fix VLM model in web_browser example docs by @albertvillanova in https://github.com/huggingface/smolagents/pull/1424 * Fix ValueError on Guided Tour docs by @albertvillanova in https://github.com/huggingface/smolagents/pull/1425 * Explain agent types better in Guided tour docs by @albertvillanova in https://github.com/huggingface/smolagents/pull/1426 * Add Quickstart page to docs by @albertvillanova in https://github.com/huggingface/smolagents/pull/1413 ## Maintenance ๐Ÿ—๏ธ * Bump dev version: v1.18.0.dev0 by @albertvillanova in https://github.com/huggingface/smolagents/pull/1390 ## New Contributors * @Lrakotoson made their first contribution in https://github.com/huggingface/smolagents/pull/1383 * @neonwatty made their first contribution in https://github.com/huggingface/smolagents/pull/1398 * @SrzStephen made their first contribution in https://github.com/huggingface/smolagents/pull/1403 * @chahn made their first contribution in https://github.com/huggingface/smolagents/pull/1414 **Full Changelog**: https://github.com/huggingface/smolagents/compare/v1.17.0...v1.18.0 ## New Features โœจ * **Structured Generation in CodeAgent**: Add optional support for structured outputs in `CodeAgent`, enabling more reliable and consistent generation patterns * Adding optional structured generation to CodeAgent by @akseljoonas in https://github.com/huggingface/smolagents/pull/1346 * **Support for Streamable HTTP MCP Servers**: Expand compatibility with new server types to support streamable HTTP MCP implementations * Support Streamable HTTP MCP servers by @albertvillanova in https://github.com/huggingface/smolagents/pull/1384 * **Run Results from `Agent.run()`**: The `Agent.run()` method can now return a `RunResult` object, providing richer metadata on agent execution * Agent.run() can return RunResult object by @aymeric-roucher in https://github.com/huggingface/smolagents/pull/1337 ## Security Enhancements ๐Ÿ”’ * **Safer LocalPythonExecutor**: Prevent indirect submodule access via attribute resolution, mitigating potential security risks in user-defined code * Prevent submodules through indirect attribute access in LocalPythonExecutor by @albertvillanova in https://github.com/huggingface/smolagents/pull/1375 ## Enhancements ๐Ÿ› ๏ธ * **LocalPythonExecutor Enhancements**: * Support annotated assignments within class definition in LocalPythonExecutor by @albertvillanova in https://github.com/huggingface/smolagents/pull/1355 * Fix evaluate_class_def for Assign Attribute target in LocalPythonExecutor by @albertvillanova in https://github.com/huggingface/smolagents/pull/1357 * Support 'pass' statement in class definition in LocalPythonExecutor by @albertvillanova in https://github.com/huggingface/smolagents/pull/1358 * Refactor tests of LocalPythonExecutor by @albertvillanova in https://github.com/huggingface/smolagents/pull/1356 * Improve LaTeX rendering in GradioUI with extended delimiter support by @albertvillanova in https://github.com/huggingface/smolagents/pull/1387 ## Bug Fixes ๐Ÿ› * **Streaming fixes** * Fix live streaming when generating planning steps by @FlorianVal in https://github.com/huggingface/smolagents/pull/1348 * Stop streaming if LiteLLM provide a finish_reason by @FlorianVal in https://github.com/huggingface/smolagents/pull/1350 * add api_base and api_keys to preparation of kwargs for generate stream by @FlorianVal in https://github.com/huggingface/smolagents/pull/1344 * Fix WebSearchTool validation error by @albertvillanova in https://github.com/huggingface/smolagents/pull/1367 * Fix smolagents benchmark by @aymeric-roucher in https://github.com/huggingface/smolagents/pull/1377 ## Documentation Improvements ๐Ÿ“š * Add example docs about using OpenRouter models by @albertvillanova in https://github.com/huggingface/smolagents/pull/1364 * Fix Llama model name in docs example by @SaiDunoyer in https://github.com/huggingface/smolagents/pull/1379 * fix typo in docstring in mcp_client.py by @grll in https://github.com/huggingface/smolagents/pull/1380 * Document use_structured_outputs_internally with version added by @albertvillanova in https://github.com/huggingface/smolagents/pull/1388 * Fix rendering of version added in docs by @albertvillanova in https://github.com/huggingface/smolagents/pull/1389 * Fix broken link in agentic RAG examples page by @vksx in https://github.com/huggingface/smolagents/pull/1363 ## Maintenance ๐Ÿ—๏ธ * Bump dev version: v1.17.0.dev0 by @albertvillanova in https://github.com/huggingface/smolagents/pull/1336 * Remove deprecated from_hf_api methods by @albertvillanova in https://github.com/huggingface/smolagents/pull/1351 ## New Contributors * @FlorianVal made their first contribution in https://github.com/huggingface/smolagents/pull/1344 * @vksx made their first contribution in https://github.com/huggingface/smolagents/pull/1363 * @akseljoonas made their first contribution in https://github.com/huggingface/smolagents/pull/1346 * @SaiDunoyer made their first contribution in https://github.com/huggingface/smolagents/pull/1379 **Full Changelog**: https://github.com/huggingface/smolagents/compare/v1.16.1...v1.17.0 ## Bug Fixes ๐Ÿ› * Revert "Add timeout mechanism for code execution" by @albertvillanova in https://github.com/huggingface/smolagents/pull/1340 **Full Changelog**: https://github.com/huggingface/smolagents/compare/v1.16.0...v1.16.1 ## New features ๐Ÿš€ * **Tools** * Add Bing search engine to WebSearchTool by @albertvillanova in https://github.com/huggingface/smolagents/pull/1313 * **Local Python executor improvements** * Support passing custom functions to local Python executor by @albertvillanova in https://github.com/huggingface/smolagents/pull/1302 * Support executor_kwargs in LocalPythonExecutor for initialization customization by @albertvillanova in https://github.com/huggingface/smolagents/pull/1314 * Add timeout mechanism for code execution by @albertvillanova in https://github.com/huggingface/smolagents/pull/1318 * **CLI improvements** * Enable local web agents via api_base and api_key by @zbloss in https://github.com/huggingface/smolagents/pull/1333 ## Important changes โš ๏ธ * **Default provider for InferenceClientModel is now "auto"** The default value of the `provider` argument in `InferenceClientModel` has changed from `"hf-inference"` to `"auto"`. With `"auto"`, the client will automatically select the first available provider for the model, following the user's configured priority at https://hf.co/settings/inference-providers * Pin minimum version of huggingface-hub to use InferenceClient auto provider by @julien-c in https://github.com/huggingface/smolagents/pull/1322 ## Bug Fixes ๐Ÿ› * **Streaming GradioUI fixes** * Fix thought yield in GradioUI for streaming and non-streaming by @albertvillanova in https://github.com/huggingface/smolagents/pull/1311 * Fix duplicate error message in GradioUI, test, and refactor by @albertvillanova in https://github.com/huggingface/smolagents/pull/1312 * Fix duplicate plan display in GradioUI when streaming by @albertvillanova in https://github.com/huggingface/smolagents/pull/1317 ## Documentation ๐Ÿ“š * Improve docs about Agentic RAG by @albertvillanova in https://github.com/huggingface/smolagents/pull/1319 * Update Readme video with streaming example by @aymeric-roucher in https://github.com/huggingface/smolagents/pull/1306 ## Misc ๐Ÿ› ๏ธ * Bump dev version: v1.16.0.dev0 by @albertvillanova in https://github.com/huggingface/smolagents/pull/1304 * Update minimum supported Gradio version to 5.14.0 to fix Sidebar-related AttributeError by @albertvillanova in https://github.com/huggingface/smolagents/pull/1309 * Fix deprecation of HfApiModel by @albertvillanova in https://github.com/huggingface/smolagents/pull/1315 * Fix CI DuckDuckGoSearchTool tests raising 202 Ratelimit by @albertvillanova in https://github.com/huggingface/smolagents/pull/1324 * Remove utils dir from Makefile check_dirs by @tobiasofsn in https://github.com/huggingface/smolagents/pull/1335 ## New Contributors * @tobiasofsn made their first contribution in https://github.com/huggingface/smolagents/pull/1335 * @zbloss made their first contribution in https://github.com/huggingface/smolagents/pull/1333 **Full Changelog**: https://github.com/huggingface/smolagents/compare/v1.15.0...v1.16.0 ## New features ๐Ÿš€ * **Streaming support** * Streaming model outputs by @aymeric-roucher in https://github.com/huggingface/smolagents/pull/1236 * Working streaming Gradio chatbot outputs by @aymeric-roucher in https://github.com/huggingface/smolagents/pull/1246 * Simplify validation of stream by @albertvillanova in https://github.com/huggingface/smolagents/pull/1242 * Fix return type hint for generate_stream and rename CompletionDelta to ChatMessageStreamDelta by @albertvillanova in https://github.com/huggingface/smolagents/pull/1247 * Remove dead code related to non-CompletionDelta streaming results by @albertvillanova in https://github.com/huggingface/smolagents/pull/1248 * **New models** * Add LiteLLM Router model by @NishantBaheti in https://github.com/huggingface/smolagents/pull/1096 * **New tools** * Implement WebSearchTool by @albertvillanova in https://github.com/huggingface/smolagents/pull/1300 * Replace DuckDuckGoSearchTool with WebSearchTool in docs by @albertvillanova in https://github.com/huggingface/smolagents/pull/1303 ## New documentation ๐Ÿ“š * Add Installation docs by @albertvillanova in https://github.com/huggingface/smolagents/pull/1293 * Add docs about Using different models: Gemini models by @albertvillanova in https://github.com/huggingface/smolagents/pull/1297 ## Fixes and improvements ๐Ÿ”ง * Fix typo in pydoc by @darxriggs in https://github.com/huggingface/smolagents/pull/1222 * Change language from semantic to lexical for BM25 by @HeardACat in https://github.com/huggingface/smolagents/pull/1230 * Move MCPClient to root-level lib and manage optional dependencies by @njbrake in https://github.com/huggingface/smolagents/pull/1238 * Update README.md by @aymeric-roucher in https://github.com/huggingface/smolagents/pull/1233 * Set \_\_name__ to "\_\_main__" in local Python interpreter by @albertvillanova in https://github.com/huggingface/smolagents/pull/1234 * Remove unused argument from the MultiStepAgent documentation by @damnever in https://github.com/huggingface/smolagents/pull/1250 * Add billing for inference providers model by @merveenoyan in https://github.com/huggingface/smolagents/pull/1260 * Fix Loading GAIA dataset by @aymeric-roucher in https://github.com/huggingface/smolagents/pull/1266 * Fix remote executor send_tools to skip pip install on empty package list by @albertvillanova in https://github.com/huggingface/smolagents/pull/1273 * Remove smolagents installation from remote executors by @albertvillanova in https://github.com/huggingface/smolagents/pull/1274 * Allow VisitWebpageTool to run on remote executors without installing smolagents by @keetrap in https://github.com/huggingface/smolagents/pull/1288 * Refactor RemotePythonExecutor.send_tools to call install_packages by @albertvillanova in https://github.com/huggingface/smolagents/pull/1281 * Remove unnecessary dependencies from DockerExecutor image by @albertvillanova in https://github.com/huggingface/smolagents/pull/1272 * Support tools with union type return by @albertvillanova in https://github.com/huggingface/smolagents/pull/1294 * Fix WikipediaSearchTool in remote executors by @albertvillanova in https://github.com/huggingface/smolagents/pull/1283 * docs: fix typo by @luisegarduno in https://github.com/huggingface/smolagents/pull/1253 * Fix Custom Final Answer Tool for ToolCallingAgent by @HeardACat in https://github.com/huggingface/smolagents/pull/1255 ## Maintenance ๐Ÿ—๏ธ * Bump dev version: v1.15.0.dev0 by @albertvillanova in https://github.com/huggingface/smolagents/pull/1221 * CI: Simplify testing by running all tests in a single command by @albertvillanova in https://github.com/huggingface/smolagents/pull/1218 * Fix deprecated typing by @albertvillanova in https://github.com/huggingface/smolagents/pull/1249 * Update Open DeepResearch requirements by @albertvillanova in https://github.com/huggingface/smolagents/pull/1265 * Lower Pillow dependency minimum version to 10.0.1 by @albertvillanova in https://github.com/huggingface/smolagents/pull/1270 * Hotfix CI: pin mcp < 1.7.0 by @albertvillanova in https://github.com/huggingface/smolagents/pull/1285 * Remove duckduckgo-search as required dependency by @albertvillanova in https://github.com/huggingface/smolagents/pull/1271 * Unpin mcp < 1.7.0 by @albertvillanova in https://github.com/huggingface/smolagents/pull/1289 * Remove markdownify as required dependency by @albertvillanova in https://github.com/huggingface/smolagents/pull/1275 ## New Contributors * @darxriggs made their first contribution in https://github.com/huggingface/smolagents/pull/1222 * @HeardACat made their first contribution in https://github.com/huggingface/smolagents/pull/1230 * @NishantBaheti made their first contribution in https://github.com/huggingface/smolagents/pull/1096 * @damnever made their first contribution in https://github.com/huggingface/smolagents/pull/1250 * @luisegarduno made their first contribution in https://github.com/huggingface/smolagents/pull/1253 **Full Changelog**: https://github.com/huggingface/smolagents/compare/v1.14.0...v1.15.0 ## New features ๐Ÿš€ * SmolAgents MCP Client by @grll in https://github.com/huggingface/smolagents/pull/1200 * Introduce a new `MCPClient` class that allows users to manage connections to one or more MCP servers, making integration with SmolAgents more flexible and scalable. * Amazon Bedrock Native API Support by @leandrodamascena in https://github.com/huggingface/smolagents/pull/1115 * Add native support for Amazon Bedrock in the `models` module, enabling seamless interaction with Bedrock-hosted models. ## Important changes ๐Ÿ’ฅ * Rename `HfApiModel` to `InferenceClientModel` by @julien-c in https://github.com/huggingface/smolagents/pull/1198 * Align the naming with its broader usage beyond the Hugging Face Hub, clarifying its purpose and usage. * Star-pattern-based import authorization by @aymeric-roucher in https://github.com/huggingface/smolagents/pull/1180 * Enhance import security by introducing a star-pattern-based authorization mechanism, allowing fine-grained control over which modules can be imported. ## Fixes and improvements ๐Ÿ”ง * Unpin pillow < 11.2.0 by @albertvillanova in https://github.com/huggingface/smolagents/pull/1140 * Pass vLLM (VLLMModel) model client params as `client_kwargs` by @sergiopaniego in https://github.com/huggingface/smolagents/pull/1137 * Docs: Add Code Agents in agency table by @aymeric-roucher in https://github.com/huggingface/smolagents/pull/1144 * Replace ChatMessage.from_hf_api with from_dict with asdict by @albertvillanova in https://github.com/huggingface/smolagents/pull/1076 * Make to_messages methods compatible with qwen vl by @aymeric-roucher in https://github.com/huggingface/smolagents/pull/1148 * Fix get_json_schema for keyword in description by @albertvillanova in https://github.com/huggingface/smolagents/pull/1146 * Support images in stream_to_gradio by @aymeric-roucher in https://github.com/huggingface/smolagents/pull/1154 * Fix: adjust PlanningStep user message to avoid azure's content filter by @bpoppinga in https://github.com/huggingface/smolagents/pull/1151 * Refactor code agent system prompt by @aymeric-roucher in https://github.com/huggingface/smolagents/pull/1153 * Fix bug authors by @aymeric-roucher in https://github.com/huggingface/smolagents/pull/1163 * Fix memory step model output in ToolCallingAgent by @keetrap in https://github.com/huggingface/smolagents/pull/1156 * Parsing model outputs: allow whitespaces after the header of code blobs by @aymeric-roucher in https://github.com/huggingface/smolagents/pull/1172 * Fix: Add default value to optional argument by @rca-umb in https://github.com/huggingface/smolagents/pull/1155 * Add more detail to code execution documentation by @maeste in https://github.com/huggingface/smolagents/pull/983 * Make TextInspectorTool serializable by @albertvillanova in https://github.com/huggingface/smolagents/pull/1176 * Clarify tool usage in agent step tutorial by @albertvillanova in https://github.com/huggingface/smolagents/pull/1175 * Provide ability to override Agent logger by @sysradium in https://github.com/huggingface/smolagents/pull/1166 * Pin mcpadapt >= 0.0.19 to include security fix by @albertvillanova in https://github.com/huggingface/smolagents/pull/1178 * Fix tool validation with multiple assignments by @albertvillanova in https://github.com/huggingface/smolagents/pull/1177 * Fix typo in the memory.mdx by @alpsencer in https://github.com/huggingface/smolagents/pull/1187 * feat: add custom Docker image support and enhanced configuration for DockerExecutor by @kingdomad in https://github.com/huggingface/smolagents/pull/1057 * Add api_key argument in HfApiModel by @aymeric-roucher in https://github.com/huggingface/smolagents/pull/947 * Implement Tool.from_dict by @albertvillanova in https://github.com/huggingface/smolagents/pull/1006 * Implement Agent.from_dict by @albertvillanova in https://github.com/huggingface/smolagents/pull/1007 * Call recursively Agent.to_dict for managed agents by @albertvillanova in https://github.com/huggingface/smolagents/pull/1008 * Support `Literal` type annotations in `@tool` for defining enums by @stephanlensky in https://github.com/huggingface/smolagents/pull/1195 * Support class docstrings in LocalPythonExecutor by @albertvillanova in https://github.com/huggingface/smolagents/pull/1192 * Reference Inference Providers more clearly by @julien-c in https://github.com/huggingface/smolagents/pull/1197 * Support annotated assignment in LocalPythonExecutor by @albertvillanova in https://github.com/huggingface/smolagents/pull/1193 * Fix evaluate_boolop for non-boolean return type by @ZeusFSX in https://github.com/huggingface/smolagents/pull/1191 * Make MultiStepAgent and abstract class by @sysradium in https://github.com/huggingface/smolagents/pull/1204 * fix: creating tool should not affect function that is being wrapped by @njbrake in https://github.com/huggingface/smolagents/pull/1203 * Remove header comments from MDX docs for proper rendering by @albertvillanova in https://github.com/huggingface/smolagents/pull/1212 * Fix CI test by updating LiteLLM error message by @albertvillanova in https://github.com/huggingface/smolagents/pull/1214 * Fix typo in evaluate_ast() docs: static_tools -> custom_tools by @JungeAlexander in https://github.com/huggingface/smolagents/pull/1217 * Make OpenDeepResearch tools serializable by @albertvillanova in https://github.com/huggingface/smolagents/pull/1206 * Add Security Policy by @albertvillanova in https://github.com/huggingface/smolagents/pull/1208 * Improve docs for stream run by @albertvillanova in https://github.com/huggingface/smolagents/pull/1215 * Skip unsupported stop parameter for reasoning models by @albertvillanova in https://github.com/huggingface/smolagents/pull/1211 ## New Contributors * @leandrodamascena made their first contribution in https://github.com/huggingface/smolagents/pull/1115 * @bpoppinga made their first contribution in https://github.com/huggingface/smolagents/pull/1151 * @rca-umb made their first contribution in https://github.com/huggingface/smolagents/pull/1155 * @alpsencer made their first contribution in https://github.com/huggingface/smolagents/pull/1187 * @stephanlensky made their first contribution in https://github.com/huggingface/smolagents/pull/1195 * @ZeusFSX made their first contribution in https://github.com/huggingface/smolagents/pull/1191 * @JungeAlexander made their first contribution in https://github.com/huggingface/smolagents/pull/1217 **Full Changelog**: https://github.com/huggingface/smolagents/compare/v1.13.0...v1.14.0 ## Important changes ๐Ÿš€ * Gradio UI now logs images observed by your agent: log observations images by @aymeric-roucher in https://github.com/huggingface/smolagents/pull/1113 * allow users to retrieve original Gradio app by @sysradium in https://github.com/huggingface/smolagents/pull/811 * Re-authorize submodule imports if top was allowed by @aymeric-roucher in https://github.com/huggingface/smolagents/pull/1103 - We had internal debate about this one: when a user authorizes import of a top level package like `additional_authorized_imports=["numpy"]`, should it: - Option 1: by default enable all subpackages like numpy.random, except potentially dangerous ones that we then have to detect through a blocklist? - Option 2: not import any subpackage, thus user have to type any subpackage they want to authorize, as in `additional_authorized_imports=["numpy", "numpy.random", "numpy.other_subpackage", ...]` Option 2 is safer, but Option 1 is much more practical, so we went with option 1. Daily disclaimer: code execution can never be 100% safe on you system, if you want to increase safety, use one of the remote executors that we propose, E2B or Docker. ## Fixes & Tweaks * Bump dev version: v1.13.0.dev0 by @albertvillanova in https://github.com/huggingface/smolagents/pull/1035 * Fix telemetry AttributeError: module 'smolagents' has no attribute 'ApiModel' by @IlyaGusev in https://github.com/huggingface/smolagents/pull/1040 * Hotfix CI: pin transformers < 4.50.0 by @albertvillanova in https://github.com/huggingface/smolagents/pull/1044 * Align name of tool __source__ by @albertvillanova in https://github.com/huggingface/smolagents/pull/1042 * Update type hints and docstrings of Agent/Tool .save with Path type by @albertvillanova in https://github.com/huggingface/smolagents/pull/1036 * Fix PythonInterperterTool description of code input with auth imports by @benediktstroebl in https://github.com/huggingface/smolagents/pull/1053 * Fix typos in Code Agent prompt by @andysalerno in https://github.com/huggingface/smolagents/pull/1063 * Validate agent name by @Misakar-0v0 in https://github.com/huggingface/smolagents/pull/768 * fix(`MLXModel`): define var to avoid `UnboundLocalError` by @SauravMaheshkar in https://github.com/huggingface/smolagents/pull/1059 * Minor fix code block rendering in Open DeepResearch README by @keetrap in https://github.com/huggingface/smolagents/pull/1055 * Add WikipediaSearchTool to default tools by @touseefahmed96 in https://github.com/huggingface/smolagents/pull/514 * Improve MLXModel call by @albertvillanova in https://github.com/huggingface/smolagents/pull/1067 * Support transformers 4.50.0 by @albertvillanova in https://github.com/huggingface/smolagents/pull/1070 * Improve Tool.push_to_hub to avoid saving to local filesystem by @albertvillanova in https://github.com/huggingface/smolagents/pull/1051 * Fix Tool.to_dict for nullable input parameter by @albertvillanova in https://github.com/huggingface/smolagents/pull/1038 * Support tool Gradio app with boolean input type by @albertvillanova in https://github.com/huggingface/smolagents/pull/1065 * Support running CI on push to ci-* branch by @albertvillanova in https://github.com/huggingface/smolagents/pull/1047 * Set Agent/Tool requirements as a sorted list by @albertvillanova in https://github.com/huggingface/smolagents/pull/1048 * Support passing raw to ChatMessage.from_dict by @albertvillanova in https://github.com/huggingface/smolagents/pull/1010 * Distinguish between AgentToolCallError and AgentToolExecutionError by @sysradium in https://github.com/huggingface/smolagents/pull/1060 * Add detail on potential threats to system by @aymeric-roucher in https://github.com/huggingface/smolagents/pull/1077 * Refactor Tool.save by @albertvillanova in https://github.com/huggingface/smolagents/pull/1075 * Fixing typos in toolcallingagent prompt by @Y-artios in https://github.com/huggingface/smolagents/pull/1082 * Refactor step yielding during streaming run to log PlanningSteps by @aymeric-roucher in https://github.com/huggingface/smolagents/pull/1083 * Make pandas a test extra dependency only by @albertvillanova in https://github.com/huggingface/smolagents/pull/1081 * Add security disclaimer to MCP docs by @albertvillanova in https://github.com/huggingface/smolagents/pull/1090 * Fix get_json_schema for docstrings with type annotations by @albertvillanova in https://github.com/huggingface/smolagents/pull/1088 * Remove transformers/huggingface-hub to check torch availability by @albertvillanova in https://github.com/huggingface/smolagents/pull/1022 * Improve ApiModel and subclasses instantiation by @albertvillanova in https://github.com/huggingface/smolagents/pull/1095 * Add trust_remote_code to ToolCollection.from_mcp by @albertvillanova in https://github.com/huggingface/smolagents/pull/1091 * Add user message after plan by @aymeric-roucher in https://github.com/huggingface/smolagents/pull/1104 * Add agent interruption by @aymeric-roucher in https://github.com/huggingface/smolagents/pull/1112 * Validate tool name as Python identifier by @albertvillanova in https://github.com/huggingface/smolagents/pull/1041 * Increase request timeout for tests that time out by @albertvillanova in https://github.com/huggingface/smolagents/pull/972 * Update README.md by @aymeric-roucher in https://github.com/huggingface/smolagents/pull/1125 * Change: expanded error exception for incorrect tool parameters by @cgthayer in https://github.com/huggingface/smolagents/pull/1099 * Hotfix CI AttributeError: 'PngImageFile' object has no attribute 'filename' by @albertvillanova in https://github.com/huggingface/smolagents/pull/1128 * Replace AgentGenerationError with AgentParsingError for ToolCallingAgent by @aymeric-roucher in https://github.com/huggingface/smolagents/pull/1131 * Fix planning_interval=1 bug in MultiStepAgent by @RolandJAAI in https://github.com/huggingface/smolagents/pull/1118 * Set main as the no-argument CLI entry point by @albertvillanova in https://github.com/huggingface/smolagents/pull/1126 ## New Contributors * @benediktstroebl made their first contribution in https://github.com/huggingface/smolagents/pull/1053 * @andysalerno made their first contribution in https://github.com/huggingface/smolagents/pull/1063 * @Misakar-0v0 made their first contribution in https://github.com/huggingface/smolagents/pull/768 * @SauravMaheshkar made their first contribution in https://github.com/huggingface/smolagents/pull/1059 * @Y-artios made their first contribution in https://github.com/huggingface/smolagents/pull/1082 * @cgthayer made their first contribution in https://github.com/huggingface/smolagents/pull/1099 **Full Changelog**: https://github.com/huggingface/smolagents/compare/v1.12.0...v1.13.0 ## What's Changed * Bump dev version: v1.12.0.dev0 by @albertvillanova in https://github.com/huggingface/smolagents/pull/982 * Fix json parsing by @sysradium in https://github.com/huggingface/smolagents/pull/987 * Fix Agent.from_hub for older smolagents exports by @albertvillanova in https://github.com/huggingface/smolagents/pull/989 * Regression fix: ChatMessageToolCall id must be a string by @sysradium in https://github.com/huggingface/smolagents/pull/990 * Always parse tool calls from APIs by @aymeric-roucher in https://github.com/huggingface/smolagents/pull/991 * Sort additional authorized imports by @keetrap in https://github.com/huggingface/smolagents/pull/998 * Fix custom prompt templates in Agent.from_hub by @albertvillanova in https://github.com/huggingface/smolagents/pull/1001 * Always parse arguments from API tool calls by @aymeric-roucher in https://github.com/huggingface/smolagents/pull/1000 * Fix TransformersModel: torch_dtype and trust_remote_code not passed to VLM by @Louis-Gv in https://github.com/huggingface/smolagents/pull/1012 * Fix double space in toolcalling_agent.yaml by @louisbrulenaudet in https://github.com/huggingface/smolagents/pull/1021 * Update images type and docstring by @keetrap in https://github.com/huggingface/smolagents/pull/1018 * Support JSON string arguments in tool calls by @albertvillanova in https://github.com/huggingface/smolagents/pull/1003 * Pass trust_remote_code to transformers tokenizer/processor by @albertvillanova in https://github.com/huggingface/smolagents/pull/1013 * Fix code using pillow required dependency by @albertvillanova in https://github.com/huggingface/smolagents/pull/1019 * Fix tests of Agent.save and Tool.save by @albertvillanova in https://github.com/huggingface/smolagents/pull/1029 * Remove dependency on transformers get_tests_dir by @albertvillanova in https://github.com/huggingface/smolagents/pull/1026 * Reduce model calls in planning_step from 2 to 1 by @aymeric-roucher in https://github.com/huggingface/smolagents/pull/985 * Bugfix: adding to prevent imaginary tool calls in CodeAgent by @IlyaGusev in https://github.com/huggingface/smolagents/pull/918 * Use pytest-datadir for shared data files by @albertvillanova in https://github.com/huggingface/smolagents/pull/1031 * Refactor Agent.to_dict by @albertvillanova in https://github.com/huggingface/smolagents/pull/1004 * Add support for MCP SSE servers by @grll in https://github.com/huggingface/smolagents/pull/931 ## New Contributors * @Louis-Gv made their first contribution in https://github.com/huggingface/smolagents/pull/1012 * @louisbrulenaudet made their first contribution in https://github.com/huggingface/smolagents/pull/1021 **Full Changelog**: https://github.com/huggingface/smolagents/compare/v1.11.0...v1.12.0 ## What's Changed * Bump dev version: v1.11.0.dev0 by @albertvillanova in https://github.com/huggingface/smolagents/pull/890 * Add Chinese translation & fix zh doc build process by @muscaestar in https://github.com/huggingface/smolagents/pull/893 * Fix chained assignment in local executor by @sysradium in https://github.com/huggingface/smolagents/pull/843 * Check dangerous modules instead of dangerous patterns by @albertvillanova in https://github.com/huggingface/smolagents/pull/877 * Fix typo in building good agents tutorial by @whistler in https://github.com/huggingface/smolagents/pull/903 * Add tests and refactor CLI by @albertvillanova in https://github.com/huggingface/smolagents/pull/892 * Bump mcpadapt >= 0.0.15 by @albertvillanova in https://github.com/huggingface/smolagents/pull/895 * Update docs about changing the system prompt by @albertvillanova in https://github.com/huggingface/smolagents/pull/900 * Support transformers 4.49.0 by @albertvillanova in https://github.com/huggingface/smolagents/pull/898 * Fix typo in list of dangerous functions by @albertvillanova in https://github.com/huggingface/smolagents/pull/909 * Improve evaluate_subscript by @albertvillanova in https://github.com/huggingface/smolagents/pull/916 * Enforce security check in all evaluations by @albertvillanova in https://github.com/huggingface/smolagents/pull/906 * Improve performance of test_while infinite loop by @albertvillanova in https://github.com/huggingface/smolagents/pull/907 * Test all dangerous functions and fix check of os.system by @albertvillanova in https://github.com/huggingface/smolagents/pull/911 * Fix Tool docstring, and model init exception by @zhanluxianshen in https://github.com/huggingface/smolagents/pull/927 * Remove default model_id from models by @albertvillanova in https://github.com/huggingface/smolagents/pull/905 * Improve docstring of parse_code_blobs by @albertvillanova in https://github.com/huggingface/smolagents/pull/876 * Test vulnerability for dangerous modules by @albertvillanova in https://github.com/huggingface/smolagents/pull/929 * Fix broken link to benchmarking code in README by @amaarora in https://github.com/huggingface/smolagents/pull/933 * Fix Gradio UI for unnamed agents by @sysradium in https://github.com/huggingface/smolagents/pull/812 * Add mlx-lm to all extra by @albertvillanova in https://github.com/huggingface/smolagents/pull/961 * Fix Gradio UI for None name by @albertvillanova in https://github.com/huggingface/smolagents/pull/956 * Introduce counter class by @Bilokin in https://github.com/huggingface/smolagents/pull/891 * Fix AzureOpenAIServerModel to not call openai.OpenAI by @albertvillanova in https://github.com/huggingface/smolagents/pull/966 * Fix CI usage of uv by @albertvillanova in https://github.com/huggingface/smolagents/pull/970 * Adds VLLMModel by @NathanHB in https://github.com/huggingface/smolagents/pull/337 * Support passing flatten_messages_as_text to OpenAIServerModel and test by @albertvillanova in https://github.com/huggingface/smolagents/pull/959 * Fix tool function implementation by @albertvillanova in https://github.com/huggingface/smolagents/pull/943 * Fix broken example of using MCP Server by @njbrake in https://github.com/huggingface/smolagents/pull/936 * Support passing any kwargs to E2BExecutor Sandbox by @aymeric-roucher in https://github.com/huggingface/smolagents/pull/951 * Fix VLLMModel example in models docs by @zhanluxianshen in https://github.com/huggingface/smolagents/pull/976 * Fix OpenAIServerModel client by @durd07 in https://github.com/huggingface/smolagents/pull/975 * Forbid all modules by default except whitelist authorized_imports by @albertvillanova in https://github.com/huggingface/smolagents/pull/935 * Forbid access to all dunder attributes by default by @albertvillanova in https://github.com/huggingface/smolagents/pull/979 * Raise agent generation errors by @aymeric-roucher in https://github.com/huggingface/smolagents/pull/948 * Update Open Deep Research's README.md by @keetrap in https://github.com/huggingface/smolagents/pull/763 * Make sure to log thoughts if verbosity_level is set to high by @aymeric-roucher in https://github.com/huggingface/smolagents/pull/964 * Get correct source code for functions defined in local executor. by @aymeric-roucher in https://github.com/huggingface/smolagents/pull/973 ## New Contributors * @muscaestar made their first contribution in https://github.com/huggingface/smolagents/pull/893 * @whistler made their first contribution in https://github.com/huggingface/smolagents/pull/903 * @zhanluxianshen made their first contribution in https://github.com/huggingface/smolagents/pull/927 * @amaarora made their first contribution in https://github.com/huggingface/smolagents/pull/933 * @NathanHB made their first contribution in https://github.com/huggingface/smolagents/pull/337 * @njbrake made their first contribution in https://github.com/huggingface/smolagents/pull/936 * @durd07 made their first contribution in https://github.com/huggingface/smolagents/pull/975 **Full Changelog**: https://github.com/huggingface/smolagents/compare/v1.10.0...v1.11.0 ## New features ๐Ÿš€๐Ÿš€ * Add Docker executor by @aymeric-roucher in https://github.com/huggingface/smolagents/pull/733 - This means you can now initialize agents with `executor="docker"` to run code in a Docker sandbox! Docs coming soon. * Safer local Python interpreter * Make local Python interpreter safer by checking if returns builtins by @albertvillanova in https://github.com/huggingface/smolagents/pull/848 * Make local Python interpreter safer by checking if returns dangerous modules by @albertvillanova in https://github.com/huggingface/smolagents/pull/861 * Make local Python interpreter safer by adding builtins to dangerous modules by @albertvillanova in https://github.com/huggingface/smolagents/pull/869 * Add support for serper search by @aymeric-roucher in https://github.com/huggingface/smolagents/pull/698 * Support running a demo of Open DeepResearch by @albertvillanova in https://github.com/huggingface/smolagents/pull/806 * Enable Colab notebook version of each doc page (by renaming doc file extension from .md to .mdx) by @albertvillanova in https://github.com/huggingface/smolagents/pull/691 ## Fixes and improvements ๐Ÿ”ง * Bump dev version: v1.10.0.dev0 by @albertvillanova in https://github.com/huggingface/smolagents/pull/649 * Improve missing toolcall exception by @GrahamTheCoder in https://github.com/huggingface/smolagents/pull/614 * refactor: simplify file type checking from MIME to extension by @kingdomad in https://github.com/huggingface/smolagents/pull/342 * Bugfix: Groq via LiteLLM ๐Ÿš… by @JGalego in https://github.com/huggingface/smolagents/pull/605 * Add docstring args for MultiStepAgent.from_folder by @albertvillanova in https://github.com/huggingface/smolagents/pull/654 * Remove non-existent subfolder arg in MultiStepAgent.from_hub by @albertvillanova in https://github.com/huggingface/smolagents/pull/653 * Minor fix save agents by @keetrap in https://github.com/huggingface/smolagents/pull/645 * Align step numbers in monitoring logging by @spemed in https://github.com/huggingface/smolagents/pull/667 * Remove deprecated resume_download arg in from_hub by @albertvillanova in https://github.com/huggingface/smolagents/pull/670 * Pass notebook_folder to build notebook docs by @albertvillanova in https://github.com/huggingface/smolagents/pull/671 * Update the prompts for delegating subtasks to agents in prompt files by @spemed in https://github.com/huggingface/smolagents/pull/678 * Documentation: manage agent's memory by @aymeric-roucher in https://github.com/huggingface/smolagents/pull/675 * Move contributing details into contributing file by @GrahamTheCoder in https://github.com/huggingface/smolagents/pull/651 * Fix typo in error message by @samee99 in https://github.com/huggingface/smolagents/pull/687 * Refactor evaluate ast to improve readability by @CalOmnie in https://github.com/huggingface/smolagents/pull/625 * Update docs in zh/examples/multiagents.md by @spemed in https://github.com/huggingface/smolagents/pull/688 * Make dangerous_patterns a module variable by @CalOmnie in https://github.com/huggingface/smolagents/pull/505 * Pin transformers < 4.49.0 to fix TypeError: LlavaProcessor: got multiple values for keyword argument 'images' by @albertvillanova in https://github.com/huggingface/smolagents/pull/693 * Respect formatting in ast.FormattedValue for local_python_executor by @LorienMan in https://github.com/huggingface/smolagents/pull/660 * Simplify instrumentation for phoenix by @aymeric-roucher in https://github.com/huggingface/smolagents/pull/697 * Test validate_tool_attributes by @albertvillanova in https://github.com/huggingface/smolagents/pull/696 * Remove translation tool from README by @keetrap in https://github.com/huggingface/smolagents/pull/705 * Fix test docs by @albertvillanova in https://github.com/huggingface/smolagents/pull/701 * Fix SpeechToTextTool by @albertvillanova in https://github.com/huggingface/smolagents/pull/706 * Fixed typo in Telemetry docs by @sergiopaniego in https://github.com/huggingface/smolagents/pull/710 * Fix validate_tool_attributes for non-literal defaults by @albertvillanova in https://github.com/huggingface/smolagents/pull/711 * Ignore parameter types in tool docstrings by @carlsberg03 in https://github.com/huggingface/smolagents/pull/724 * Refactor MultiStepAgent and improve some tests by @colesmcintosh in https://github.com/huggingface/smolagents/pull/598 * Fix duplicate managed agent name and its tool name by @keetrap in https://github.com/huggingface/smolagents/pull/721 * Update docs on how to change prompts by @albertvillanova in https://github.com/huggingface/smolagents/pull/725 * Remove transformers tasks from load_tool docstring by @albertvillanova in https://github.com/huggingface/smolagents/pull/726 * Remove article section from launch_gradio_demo Interface by @albertvillanova in https://github.com/huggingface/smolagents/pull/728 * Update logo in Readme by @aymeric-roucher in https://github.com/huggingface/smolagents/pull/730 * Support passing additional parameters for openai client initialization by @touseefahmed96 in https://github.com/huggingface/smolagents/pull/388 * Add api-base and api-key arguments to CLI by @keetrap in https://github.com/huggingface/smolagents/pull/734 * Fix CI test error: DuckDuckGoSearchException: 202 Ratelimit by @albertvillanova in https://github.com/huggingface/smolagents/pull/739 * Fix typo in Makefile by @albertvillanova in https://github.com/huggingface/smolagents/pull/741 * Add disclaimer in benchmark to request access for gated dataset by @albertvillanova in https://github.com/huggingface/smolagents/pull/744 * Define test RUN_ALL requirement marker in one file by @albertvillanova in https://github.com/huggingface/smolagents/pull/743 * Rename managed agent in multiagents docs to avoid confusing the LLM by @albertvillanova in https://github.com/huggingface/smolagents/pull/766 * Update documentation on sandboxes by @aymeric-roucher in https://github.com/huggingface/smolagents/pull/747 * Add audioop-lts as Python>=3.13 Open DeepResearch requirement for pydub by @albertvillanova in https://github.com/huggingface/smolagents/pull/749 * Create docker extra installation dependency by @albertvillanova in https://github.com/huggingface/smolagents/pull/764 * Fix bracket logging in rich by @aymeric-roucher in https://github.com/huggingface/smolagents/pull/753 * Enable custom final answer in CodeAgent by @aymeric-roucher in https://github.com/huggingface/smolagents/pull/769 * Fix TypeError raised by launch_gradio_demo with PipelineTool by @albertvillanova in https://github.com/huggingface/smolagents/pull/732 * Use docker extra in Secure code execution docs by @albertvillanova in https://github.com/huggingface/smolagents/pull/780 * Support custom final_answer in agent init by @albertvillanova in https://github.com/huggingface/smolagents/pull/783 * Fix LiteLLMModel API key usage in CLI by @asselindebeauville in https://github.com/huggingface/smolagents/pull/788 * Secure code execution wrong title in docs TOC by @maeste in https://github.com/huggingface/smolagents/pull/771 * Update Open DeepResearch to install smolagents from local path by @albertvillanova in https://github.com/huggingface/smolagents/pull/786 * Support importing scipy and sklearn without AttributeError: module 'scipy.sparse._coo' has no attribute 'upcast' by @albertvillanova in https://github.com/huggingface/smolagents/pull/790 * Fix and test validate_tools_and_managed_agents by @albertvillanova in https://github.com/huggingface/smolagents/pull/731 * Update README with mention to Docker sandboxed execution by @touseefahmed96 in https://github.com/huggingface/smolagents/pull/803 * Added sidebar layout and session state management for agent by @yvrjsharma in https://github.com/huggingface/smolagents/pull/797 * Fix CI quality in examples/gradio_ui by @albertvillanova in https://github.com/huggingface/smolagents/pull/822 * Improve dependency loading of Open DeepResearch visual_qa by @albertvillanova in https://github.com/huggingface/smolagents/pull/821 * Improve setup instruction for Open DeepResearch by @albertvillanova in https://github.com/huggingface/smolagents/pull/823 * Support other models than o1 in Open Deep Research by @albertvillanova in https://github.com/huggingface/smolagents/pull/824 * Fix final_answer_pattern matching in remote executors by @albertvillanova in https://github.com/huggingface/smolagents/pull/826 * Fix and refactor tests of Docker executors by @albertvillanova in https://github.com/huggingface/smolagents/pull/827 * Remove unnecessary self argument in GoogleSearchTool super() call by @gael1130 in https://github.com/huggingface/smolagents/pull/828 * Fix call from dict with LocalPythonExecutor by @albertvillanova in https://github.com/huggingface/smolagents/pull/837 * Minor fix of import in VisitWebpageTool by @keetrap in https://github.com/huggingface/smolagents/pull/840 * No need of these Files by @touseefahmed96 in https://github.com/huggingface/smolagents/pull/798 * fix: correcting tool_call role by @zndqs in https://github.com/huggingface/smolagents/pull/779 * Fix translation guide as it points to invalid location by @FanwangM in https://github.com/huggingface/smolagents/pull/794 * Fix load_model in webagent CLI by @albertvillanova in https://github.com/huggingface/smolagents/pull/855 * Fix message for observations without tool_calls by @albertvillanova in https://github.com/huggingface/smolagents/pull/857 * Fix CI test_action_step_to_messages by @albertvillanova in https://github.com/huggingface/smolagents/pull/868 * Fix LiteLLMModel docstring by @albertvillanova in https://github.com/huggingface/smolagents/pull/885 * Fix bug where operations count was not reinitialized at each step by @aymeric-roucher in https://github.com/huggingface/smolagents/pull/866 * Fix CLI webagent by @albertvillanova in https://github.com/huggingface/smolagents/pull/889 ## New Contributors * @JGalego made their first contribution in https://github.com/huggingface/smolagents/pull/605 * @spemed made their first contribution in https://github.com/huggingface/smolagents/pull/667 * @samee99 made their first contribution in https://github.com/huggingface/smolagents/pull/687 * @LorienMan made their first contribution in https://github.com/huggingface/smolagents/pull/660 * @carlsberg03 made their first contribution in https://github.com/huggingface/smolagents/pull/724 * @colesmcintosh made their first contribution in https://github.com/huggingface/smolagents/pull/598 * @asselindebeauville made their first contribution in https://github.com/huggingface/smolagents/pull/788 * @maeste made their first contribution in https://github.com/huggingface/smolagents/pull/771 * @yvrjsharma made their first contribution in https://github.com/huggingface/smolagents/pull/797 * @gael1130 made their first contribution in https://github.com/huggingface/smolagents/pull/828 * @zndqs made their first contribution in https://github.com/huggingface/smolagents/pull/779 **Full Changelog**: https://github.com/huggingface/smolagents/compare/v1.9.2...v1.10.0 ## Bug Fixes ๐Ÿ› * Fix litellm flatten_messages_as_text detection by @aymeric-roucher in https://github.com/huggingface/smolagents/commit/ef6031542c1f71b758e3167b9c21888a726f92d8 * Fix litellm flatten_messages_as_text detection by @aymeric-roucher in https://github.com/huggingface/smolagents/pull/659 **Full Changelog**: https://github.com/huggingface/smolagents/compare/v1.9.1...v1.9.2