What's Changed

• Bump dev version: v1.26.0.dev0 by @albertvillanova in #2273
• Add new possible error message to test of LiteLLM providers w/o key by @albertvillanova in #2275
• Fix typo cleanups across docs, examples, and tests by @luojiyin1987 in #2208
• fix: correct f-string escaping in ToolSerializer.dumps error message by @Ricardo-M-L in #2134
• Revert fix: correct f-string escaping by @albertvillanova in #2291
• Update docstring of LocalPythonExecutor by @albertvillanova in #2292
• Add Exa as a search engine option in WebSearchTool by @10ishq in #2139
• Remove remote WasmExecutor by @albertvillanova in #2321

New Contributors

• @luojiyin1987 made their first contribution in #2208
• @Ricardo-M-L made their first contribution in #2134
• @10ishq made their first contribution in #2139

Full Changelog: v1.25.0...v1.26.0

What's Changed

• Bump dev version: v1.25.0.dev0 by @albertvillanova in #1939
• Fix import of TokenUsage by @albertvillanova in #1943
• Add MLflow integration doc by @B-Step62 in #1884
• Refactor agent/model deserialization: Replace importlib with registry pattern by @albertvillanova in #1944
• Update _function_type_hints_utils.py: CohereForAI -> CohereLabs by @Michellehbn in #1946
• Fix High impact vulnerability on remote executors by @nnfrog in #1637
• Fix SafeSerializer.get_deserializer_code by @albertvillanova in #1983
• Fix RemotePythonExecutor._deserialize_final_answer for legacy no-prefix pickle by @albertvillanova in #1984
• Test 'pickle:...' payloads are properly decoded by @albertvillanova in #1985
• Remove support for legacy no-prefix pickle by @albertvillanova in #1986
• Create docs reference page for Python executors by @albertvillanova in #1987
• Isolate WasmExecutor Deno cache to per-instance temp directory by @albertvillanova in #1995
• Switch wildcard bind to loopback-only endpoint in WasmExecutor by @albertvillanova in #2025
• Use token in DockerExecutor by @albertvillanova in #2027
• Remove allow_origin from Docker and Modal executors by @albertvillanova in #2028
• Use token in WasmExecutor by @albertvillanova in #2031
• Fix evaluate_with() calling exit on wrong object by @dubin555 in #2029
• Fix evaluate_with by suppressing exception if truthy exit by @albertvillanova in #2033
• Use requests.Session in WasmExecutor by @albertvillanova in #2032
• Fix Rich MarkupError in AgentLogger.log_task by rendering untrusted text safely by @Barcavin in #1894
• Emphasise sandboxing is required; LocalPythonExecutor is not a security tool by @zsims in #2039
• Fix log_error rendering brackets as Rich markup by @albertvillanova in #2077
• Support huggingface-hub >= 1 by @merveenoyan in #2075
• Remove unused safe_serialization param from _websocket_run_code_raise_errors by @albertvillanova in #2131
• Set allow_pickle default to False in internal calls by @albertvillanova in #2132
• 🔒 Pin GitHub Actions to commit SHAs by @paulinebm in #2149
• chore: bump doc-builder SHA for PR upload workflow by @rtrompier in #2194
• chore: bump doc-builder SHA for main doc build workflow by @rtrompier in #2219

New Contributors

• @B-Step62 made their first contribution in #1884
• @Michellehbn made their first contribution in #1946
• @dubin555 made their first contribution in #2029
• @zsims made their first contribution in #2039
• @paulinebm made their first contribution in #2149
• @rtrompier made their first contribution in #2194

Full Changelog: v1.24.0...v1.25.0

What's Changed

• Bump version to 1.24 by @aymeric-roucher in #1871
• Fix: Add backward compatibility for deprecated HfApiModel by @MohammadKassas143 in #1863
• Fix typo in secure_code_execution.md by @jonathanagustin in #1892
• Update no-stop-sequence model list to support gpt-5.2* by @aymeric-roucher in #1895
• Fix run_gaia.py token_counts when managed agent is called more than once by @suryabdev in #1878
• 🌐 [i18n-KO] Translated tools.md to Korean by @Kim-Ju-won in #1835
• Support passing additional params to apply_chat_template by @albertvillanova in #1902
• fix(gradio_ui): Gradio 6 compatibility by @alejandro-ao in #1903
• Coerce tool calls from APIs into smolagents ChatMessageToolCall format by @aymeric-roucher in #1900
• Add FinalAnswerStep to possible step_callbacks by @suryabdev in #1881
• Upgrade GitHub Actions for Node 24 compatibility by @salmanmkc in #1909
• fix: prevent FinalAnswerException from being caught by except Exception by @majiayu000 in #1914
• Add DeepWiki badge to README by @aymeric-roucher in #1846
• Implement robust timeout mechanism for Python execution by @albertvillanova in #1910

New Contributors

• @MohammadKassas143 made their first contribution in #1863
• @jonathanagustin made their first contribution in #1892
• @alejandro-ao made their first contribution in #1903
• @salmanmkc made their first contribution in #1909
• @majiayu000 made their first contribution in #1914

Full Changelog: v1.23.0...v1.24.0

Improvements ✨

• Update final answer checks to accept the agent instance by @Zoe14 in #1648
• Dialog mode CLI by @aymeric-roucher in #1540
• Add Blaxel support for remote code execution by @Joffref in #1791
• Proposition: Add exponential backoff with jitter for retries by @suryabdev in #1829
• Cut generation at stop sequence post generation by @aymeric-roucher in #1802
• Move executor type validation to python executor creation by @Lrakotoson in #1800
• Support custom Python code executor in CodeAgent by @albertvillanova in #1707
• Retry on rate limit errors by @aymeric-roucher in #1801
• Change default InferenceClient model to Qwen/Qwen3-Next-80B-A3B-Thinking by @suryabdev in #1813
• Remove 'Server' from model names in doc by @aymeric-roucher in #1812
• Optimize comprehension evaluation with generator-based approach in LocalPythonExecutor by @albertvillanova in #1824
• LocalPythonExecutor: Add support for nested dictcomp and setcomp by @suryabdev in #1823
• Support parsing anyOf from MCP tools by @duguyue100 in #1708
• Add Modal as sandbox option to docs front page by @sandcat100 in #1865
• make additional_args nullable for managed agents by @nflo in #1827
• vLLM: Move from guided_options_request to structured_outputs by @suryabdev in #1805

Fixes 🛠️

• Fix agent logs: logging "None" and simplify the logic by @albertvillanova in #1788
• Fix dict(message) bug in AgentMemory.replay for ChatMessage objects by @Flakes342 in #1763
• Fix LocalPythonExecutor support for Enum and other metaclasses by @colesmcintosh in #1797
• Handle None content in stop-sequence trimming by @chahn in #1826
• Fix minor benchmark script bugs by @suryabdev in #1822
• Fix YAML scanner error when building documentation by @albertvillanova in #1841
• Pin huggingface-hub <1.0.0 to fix failed to build tokenizers by @albertvillanova in #1844
• Fixed SyntaxError format: keep exception type and message on same line (#835) by @balazsthomay in #1864
• Fix CLI Tool.from_space() call by auto-generating name and description (#1535) by @balazsthomay in #1859
• Add gpt-5.1 support by @suryabdev in #1868
• bug: role not converted back to enum by @njbrake in #1833
• Parallel tool calls lose OpenTelemetry context: tool spans not nested under agent step by @Barcavin in #1840

Documentation 📚

• Bump dev version: v1.23.0.dev0 by @albertvillanova in #1787
• 🌐 [i18n-KO] Translated memory.md to Korean by @HyunZ118 in #1793
• 🌐 [i18n-KO] Translated using_different_models.md to Korean by @FacerAin in #1772
• 🌐 [i18n-KO] Translated guided_tour.md to Korean by @FacerAin in #1711
• 🌐 [i18n-KO] Translated async_agent.md to Korean by @FacerAin in #1749
• 🌐 [i18n-KO] Translated agents.md to Korean by @Kim-Ju-won in #1720
• 🌐 [i18n-KO] Translated rag.md to Korean by @Kim-Ju-won in #1646
• Documentation: Minor fixes by @suryabdev in #1809
• 🌐 [i18n-KO] Translated web_browser.md to Korean by @FacerAin in #1748
• [i18n-es] Translating docs to spanish by @1-echo in #1806
• 🌐 [i18n-KO] Translated plan_customization.md to Korean by @FacerAin in #1771
• 🌐 [i18n-KO] Translated inspect_runs.md to Korean by @FacerAin in #1747
• Build PR documentation for all languages by @albertvillanova in #1842

Full Changelog: v1.22.0...v1.23.0

What's Changed

• Bump dev version: v1.22.0.dev0 by @albertvillanova in #1657
• refactor(PythonExecutor): inherit from ABC and declare abstract methods by @HairlessVillager in #1654
• Fix error for GPT-5 not supporting the stop parameter by @Xmader in #1662
• Enable return_full_result directly in run method by @aymeric-roucher in #1658
• Make RunResult json serializable by @aymeric-roucher in #1659
• Change truncation to remove end rather than middle by @aymeric-roucher in #1665
• Document that model kwargs are forwarded to model completion call by @albertvillanova in #1679
• Fix missing kwargs forward to vLLM model generate call by @albertvillanova in #1680
• Raise RuntimeError with LiteLLM response details if no choices by @albertvillanova in #1683
• Fix AmazonBedrockModel with reasoning/thinking content by @albertvillanova in #1681
• Fix installation command in README by quoting extra by @shahgahmed in #1685
• CI hotfix: Pin openai < 1.100.0 for litellm extra by @albertvillanova in #1693
• Support openai 1.100 for litellm extra by @albertvillanova in #1698
• Pin telemetry extra >=0.1.15 to use new TokenUsage structure by @albertvillanova in #1699
• Add AGENTS.md by @albertvillanova in #1701
• Test telemetry extra by @albertvillanova in #1700
• Replace slim with bullseye base image in Docker executor by @albertvillanova in #1702
• fix: error missing f-string by @njbrake in #1691
• Support custom Dockerfile in DockerExecutor by @albertvillanova in #1705
• Support MCP structured output and output schema by @chahn in #1663
• Improve agent image handling by @aaron-ang in #1601
• 🌐 [i18n-KO] Translated multiagents.md to Korean by @pyapyapya in #1644
• Fix CI 403 error for Wikipedia page in test_visit_webpage by @albertvillanova in #1716
• fix typos in the system prompt by @Frame17 in #1710
• Fix pip install commands by quoting extras by @albertvillanova in #1690
• Fix deprecation warning about parameter messages by @albertvillanova in #1718
• Fix warning about parameter max_new_tokens by @albertvillanova in #1721
• Adds Modal Remote Executor by @thomasjpfan in #1722
• Fix formatting of title in translation section of documentation by @albertvillanova in #1727
• Remove untranslated sections from the documentation by @albertvillanova in #1729
• fix: add ipykernel in dockerfile of prebuilt docker executor by @cyyeh in #1731
• Fix E2BExecutor by pinning e2b-code-interpreter to < 2 by @albertvillanova in #1733
• Fix DockerExecutor connection reset error with server readiness check by @benoriol in #1684
• [i18n-KO] Translated building_good_agents.md to Korean by @FacerAin in #1714
• 🌐 [i18n-KO] Translated installation.md to Korean by @FacerAin in #1647
• Support e2b-code-interpreter v2 by @August-murr in #1751
• Fix token count initialization when plan_message.token_usage is None by @satwikkansal in #1752
• Support user-configurable parameter override for model completion parameters by @albertvillanova in #1678
• Fix error for grok-4/grok-3-mini not supporting the stop parameter by @suryabdev in #1766
• Replace DockerExecutor Dockerfile with file object by @albertvillanova in #1738
• Update documentation to clarify instructions are appended by @chahn in #1709
• 🌐 [i18n-KO] Translated react.md to Korean by @NSBG in #1666
• Pass tools_to_call_from to TransformersModel._prepare_completion_kwargs by @albertvillanova in #1782
• 🌐 [i18n-KO] Translated models.md to Korean by @Kim-Ju-won in #1776
• Fix AttributeError when trying to log a None by @albertvillanova in #1786
• fix(agent): remove unused get_variable_names function by @keli-wen in #1784
• Fix XPath injection in search_item_ctrl_f by @albertvillanova in #1768
• Use context manager for per-operation websocket connection in DockerExecutor by @albertvillanova in #1750

New Contributors

• @Xmader made their first contribution in #1662
• @shahgahmed made their first contribution in #1685
• @aaron-ang made their first contribution in #1601
• @pyapyapya made their first contribution in #1644
• @Frame17 made their first contribution in #1710
• @cyyeh made their first contribution in #1731
• @satwikkansal made their first contribution in #1752
• @suryabdev made their first contribution in #1766
• @Kim-Ju-won made their first contribution in #1776
• @keli-wen made their first contribution in #1784

Full Changelog: v1.21.3...v1.22.0

Bug Fixes 🐛

• Fix E2BExecutor by pinning e2b-code-interpreter to < 2 by @albertvillanova in #1733

Full Changelog: v1.21.2...v1.21.3

Bug Fixes 🐛

• Pin telemetry extra >=0.1.15 to use new TokenUsage structure by @albertvillanova in #1699

Full Changelog: v1.21.1...v1.21.2

Bug Fixes 🐛

• Fix error for GPT-5 not supporting the stop parameter by @Xmader in #1662

Full Changelog: v1.21.0...v1.21.1

Enhancements ✨

• Tool & Model Improvements

  • Add Tool to_code_prompt and to_tool_calling_prompt by @albertvillanova in #1599
  • Improve function body extraction in tools by @HairlessVillager in #1627
  • Support passing model_kwargs to TransformersModel by @albertvillanova in #1608
  • Support dict messages as direct input to models by @albertvillanova in #1570

• Executor Improvements

  • Enhance LocalPythonExecutor security by blocking dunder calls by @nnfrog in #1551
  • Fix continue semantics of LocalPythonExecutorby @adaminsky in #1645
  • Fix GeneratorExp in LocalPythonExecutor to return Generator instead of list by @albertvillanova in #1650

• Gradio & UI

  • Reset agent memory on GradioUI clear button by @albertvillanova in #1611
  • Use gr.Number for mapping integer and number type components in launch_gradio_demo by @karenwky in #1320

Bug Fixes 🐛

• Fix: Prevent empty additional_args from being added to agent prompt by @jokokojote in #1566
• Fix SyntaxError in generated Gradio app due to 'class' attribute by @albertvillanova in #1615
• Fix serialization inconsistency in MultiStepAgent.to_dict/from_dict for managed_agents by @hazrmard in #1577
• Fix a kwargs error in LocalPythonExecutor by @aymeric-roucher in #1583
• Fix structured outputs by @MicrosoftJames in #1642
• Fix bug where AmazonBedrockServerModel crashes in thinking mode by @benoriol in #1632
• Fix: Remove duplicate line in tool_calling_agent prompt by @yusuf-eren in #1636
• Fix: RemotePythonExecutor should not send variables if none exist by @thomasjpfan in #1605
• Disable parallel tool calls for final answer by @aymeric-roucher in #1539

Documentation 📚

• Improve Installation docs with uv option by @albertvillanova in #1550
• Fix and improve MCP server security warnings in docs by @albertvillanova in #1549
• DOC: Fix image link for docs (secure_code_execution.md) by @FacerAin in #1582
• Update title of page "What are agents" by @aymeric-roucher in #1546
• Enable multilingual docs by @albertvillanova in #1633
• 🌐 [i18n-KO] Translated index.md to Korean by @FacerAin in #1581
• 🌐 [i18n-KO] Translated text_to_sql.md to Korean by @NSBG in #1603

CI / Dependency Updates 🧪

• Pin mcpadapt to support image/audio content by @albertvillanova in #1558
• Temporarily skip wrongly implemented tests by @albertvillanova in #1571
• CI hotfix: Pin mlx < 0.26.5 by @albertvillanova in #1586
• Support mlx-0.26.5 by @albertvillanova in #1589
• CI hotfix: Pin transformers < 4.54.0 by @albertvillanova in #1620
• Support transformers 4.54.0 by @albertvillanova in #1625
• Fix CI PytestUnknownMarkWarning: Unknown pytest.mark.timeout by @albertvillanova in #1630
• Fix CI quality: remove trailing whitespace by @albertvillanova in #1617

Deprecations & Cleanups 🧹

• Replace deprecated ast.Num and ast.Str with ast.Constant in tool validation by @WilliamRabuel in #1594
• Remove deprecated grammar parameter by @albertvillanova in #1651
• Remove deprecated token count attributes by @albertvillanova in #1652
• Remove deprecated agent logs attribute by @albertvillanova in #1653
• Remove mention of deprecated ManagedAgent from docs by @albertvillanova in #1655
• Remove deprecated default sse transport by @albertvillanova in #1656
• Replace duckduckgo_search with ddgs to address package rename warning by @WilliamRabuel in #1593
• Remove unused is_same_item function by @albertvillanova in #1631
• Remove unnecessary import by @rahul-1729 in #1639
• Simplify _is_package_available by @albertvillanova in #1641

Miscellaneous 🔧

• Add support for Amazon Bedrock API Keys by @colesmcintosh in #1612
• Factorize create_agent_gradio_app_template function by @albertvillanova in #1616
• Fix: Allow passing integers for floats by @aymeric-roucher in #1623
• Fix Tool.forward to raise NotImplementedError instead of returning by @HairlessVillager in #1628
• Make validate_tool_arguments raise instead of returning error string by @albertvillanova in #1634
• Remove mention of python in additional args by @aymeric-roucher in #1584
• ToolCallingAgent: remove unnecessary addition to model_output by @aymeric-roucher in #1622
• Minor fix in toolCallingAgentPrompt by @rahul-1729 in #1574

Maintenance 🏗️

• Bump dev version: v1.21.0.dev0 by @albertvillanova in #1544
• Remove empty issue template by @albertvillanova in #1562
• Improve issue templates by @albertvillanova in #1563
• Add issue template for documentation improvements by @albertvillanova in #1564
• Add uv option to CONTRIBUTING by @albertvillanova in #1561

New Contributors

• @jokokojote made their first contribution in #1566
• @rahul-1729 made their first contribution in #1574
• @WilliamRabuel made their first contribution in #1594
• @FacerAin made their first contribution in #1582
• @thomasjpfan made their first contribution in #1605
• @hazrmard made their first contribution in #1577
• @HairlessVillager made their first contribution in #1628
• @benoriol made their first contribution in #1632
• @yusuf-eren made their first contribution in #1636
• @nnfrog made their first contribution in #1551
• @NSBG made their first contribution in #1603
• @karenwky made their first contribution in #1320
• @MicrosoftJames made their first contribution in #1642
• @adaminsky made their first contribution in #1645

Full Changelog: v1.20.0...v1.21.0

New Features ✨

• Remote Execution:

  • Implement remote Python WasmExecutor by @albertvillanova in #1261

• Callbacks:

  • Support callbacks after planning step by @albertvillanova in #1445
  • Support multiple callbacks per step type in step_callbacks dict by @albertvillanova in #1517

Enhancements 🛠️

• Models:

  • Implement rate limiting in all API models by @albertvillanova in #1516

• Tools:

  • Implement exception mechanism for final answer in remote executors by @tobiasofsn in #1429
  • Set default api_key_name in ApiWebSearchTool by @albertvillanova in #1507
  • Implement rate limit mechanism in ApiWebSearchTool by @albertvillanova in #1508
  • Implement rate limit mechanism in DuckDuckGoSearchTool by @albertvillanova in #1515
  • Enable image output for Tool.from_space by @aymeric-roucher in #1510
  • Add adapter_kwargs parameter to MCPClient by @LucasDltg in #1528

• Validation of Tool Arguments:

  • Support multiple types in tool arguments validation by @duguyue100 in #1505
  • Fix input type validation when input schema has multiple types by @duguyue100 in #1530
  • Fix optional args toolcalling by @aymeric-roucher in #1538

• Agent Output Formatting:

  • Implement CodeOutput as analog to ToolOutput by @albertvillanova in #1496
  • Allow markdown or custom formatting for code blocks by @aymeric-roucher in #1493

• Other:

  • Add ApiWebSearchTool to all list by @albertvillanova in #1492
  • Enforced type annotations in ChatMessage via MessageRole enum by @daavoo in #1479
  • Check authorized imports in LocalPythonExecutor constructor by @jank in #265

Bug Fixes 🐛

• Fix KeyError when agent reaches max steps with image input by @robin-ede in #1525
• Fix when response.usage is None for some OpenRouter models by @sentienthouseplant in #1402
• Fix incorrect token counting in streaming TransformersModel by @albertvillanova in #1503
• Minor fix nested quotes that prevent execution by @vinoo-ganesh-keru in #1497
• Fix ChatMessage attribute access after replacing dict with dataclass by @albertvillanova in #1534
• Fix access of content field in ChatMessage by @mdr223 in #1533
• Fix parallel final answers handling by @aymeric-roucher in #1482

Documentation Improvements 📚

• Examples & Tutorials:

  • Add complete Plan Customization example by @albertvillanova in #1506
  • Add Async App with Agent example by @albertvillanova in #1509
  • Improve Plan Customization example with Human-in-the-Loop by @albertvillanova in #1513

• API References:

  • Fix docstring of WikipediaSearchTool by @albertvillanova in #1521
  • Explain rate_limit in tools docstrings by @albertvillanova in #1519
  • Fix Examples in AmazonBedrockServerModel docstring by @albertvillanova in #1522
  • Split Built-in Tools from Tools in Reference docs by @albertvillanova in #1520

• Other:

  • Update Langfuse tracing documentation to SDK v3 by @jannikmaierhoefer in #1447
  • Fix use of context manager agent in docs by @albertvillanova in #1502

Maintenance 🏗️

• Fix CI LiteLLM test_call_different_providers_without_key by @albertvillanova in #1527
• Fix test_send_tools_with_default_wikipedia_search_tool by @albertvillanova in #1494
• Fix DockerExecutor tests with final_answer by calling send_tools by @albertvillanova in #1495
• Bump dev version: v1.20.0.dev0 by @albertvillanova in #1477

New Contributors

• @sentienthouseplant made their first contribution in #1402
• @daavoo made their first contribution in #1479
• @duguyue100 made their first contribution in #1505
• @vinoo-ganesh-keru made their first contribution in #1497
• @robin-ede made their first contribution in #1525
• @LucasDltg made their first contribution in #1528
• @mdr223 made their first contribution in #1533

Full Changelog: v1.19.0...v1.20.0