v1.20.0

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

v1.19.0

Enhancements 🛠️

• Agent Upgrades:

  • Support managed agents in ToolCallingAgent by @albertvillanova in #1456
  • Support context managers for agent cleanup by @tobiasofsn in #1422
  • Change code tags to xml by @aymeric-roucher in #1442

• UI Improvements:

  • Support reset_agent_memory in GradioUI by @JakeBx in #1420

• Streaming Refactor:

  • Transfer aggregation of streaming events off the Model class by @aymeric-roucher in #1449

• Agent Output Tracking:

  • Store CodeAgent code outputs in ActionStep by @aymeric-roucher in #1463

Bug Fixes 🐛

• Fix Agent update planning logic by @Zoe14 in #1417
• Remove plural from named argument return_full_results in examples by @vladlen32230 in #1434
• Fix and refactor final answer checks by @aymeric-roucher in #1448
• Fix logging of Docker build logs by @tobiasofsn in #1421
• Add a mention of additional_args in the manager agent's prompt by @aymeric-roucher in #1441
• Fix missing mentions of additional_args in manager agent prompts by @albertvillanova in #1459
• Fix: __new__ method only accepts class object, remove **args and **kw… by @abdulhakkeempa in #1462
• Do not wrap types in safer_func by @albertvillanova in #1475
• Match multiline final answers in remote executors by @albertvillanova in #1444
• Revert removal of the last message from memory_messages during planning by @Zoe14 in #1454

Documentation Improvements 📚

• Quickstart Enhancements:

  • Add simple tool usage in doc quickstart by @aymeric-roucher in #1470
  • Add open-colab link to Quickstart docs by @albertvillanova in #1472

• Reference Docs:

  • Add AgentMemory to Reference docs by @albertvillanova in #1473
  • Add docstrings to GradioUI by @albertvillanova in #1451

• Fixes & Formatting:

  • Correct broken link in documentation homepage by @johntony366 in #1468
  • Rename docs files from .mdx to .md by @albertvillanova in #1471

Maintenance 🏗️

• Bump dev version: v1.19.0.dev0 by @albertvillanova in #1427

New Contributors

• @Zoe14 made their first contribution in #1417
• @vladlen32230 made their first contribution in #1434
• @JakeBx made their first contribution in #1420
• @abdulhakkeempa made their first contribution in #1462
• @johntony366 made their first contribution in #1468

Full Changelog: v1.18.0...v1.19.0

v1.18.0

New Features ✨

• 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 #1412
• Streaming Output for ToolCallingAgent: ToolCallingAgent now supports streaming outputs, improving responsiveness and user experience during multi-step tool interactions
  • Streaming outputs for ToolCallingAgent 🚀 by @aymeric-roucher in #1409
• API Web Search Tool: Introduced a new ApiWebSearchTool, enabling structured web search capabilities via API.
  • Create API web search tool by @albertvillanova in #1400

Enhancements 🛠️

• Update prompts to avoid confusion: search -> web_search and wiki -> wiki_search by @SrzStephen in #1403
• Make Agent.system_prompt read only by @albertvillanova in #1399
• Support configurable tool_choice in prepare_completion_kwargs by @albertvillanova in #1392
• Support passing additional params to MLXModel load and tokenizer.apply_chat_template by @albertvillanova in #1406
• Support custom headers/params for ApiWebSearchTool by @albertvillanova in #1411

Bug Fixes 🐛

• Fix: Support custom inputs execution for custom Final Answer Tool by @Lrakotoson in #1383
• Fix @tool decorator for remote Python executor by @tobiasofsn in #1334
• Always pass add_generation_prompt=True to apply_chat_template by @albertvillanova in #1416

Documentation Improvements 📚

• Make docs neutral about agent types by @julien-c in #1376
• Add MCP tools doc section by @aymeric-roucher in #1394
• Update ToolCollection.from_mcp docstring examples by @neonwatty in #1398
• Document better final_answer_checks by @albertvillanova in #1407
• Fix typo in secure code execution documentation by @chahn in #1414
• Add chat server demo to examples by @albertvillanova in #1415
• Fix VLM model in web_browser example docs by @albertvillanova in #1424
• Fix ValueError on Guided Tour docs by @albertvillanova in #1425
• Explain agent types better in Guided tour docs by @albertvillanova in #1426
• Add Quickstart page to docs by @albertvillanova in #1413

Maintenance 🏗️

• Bump dev version: v1.18.0.dev0 by @albertvillanova in #1390

New Contributors

• @Lrakotoson made their first contribution in #1383
• @neonwatty made their first contribution in #1398
• @SrzStephen made their first contribution in #1403
• @chahn made their first contribution in #1414

Full Changelog: v1.17.0...v1.18.0

v1.17.0

New Features ✨

• 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 #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 #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 #1337

Security Enhancements 🔒

• 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 #1375

Enhancements 🛠️

• LocalPythonExecutor Enhancements:
  • Support annotated assignments within class definition in LocalPythonExecutor by @albertvillanova in #1355
  • Fix evaluate_class_def for Assign Attribute target in LocalPythonExecutor by @albertvillanova in #1357
  • Support 'pass' statement in class definition in LocalPythonExecutor by @albertvillanova in #1358
  • Refactor tests of LocalPythonExecutor by @albertvillanova in #1356
• Improve LaTeX rendering in GradioUI with extended delimiter support by @albertvillanova in #1387

Bug Fixes 🐛

• Streaming fixes
  • Fix live streaming when generating planning steps by @FlorianVal in #1348
  • Stop streaming if LiteLLM provide a finish_reason by @FlorianVal in #1350
  • add api_base and api_keys to preparation of kwargs for generate stream by @FlorianVal in #1344
• Fix WebSearchTool validation error by @albertvillanova in #1367
• Fix smolagents benchmark by @aymeric-roucher in #1377

Documentation Improvements 📚

• Add example docs about using OpenRouter models by @albertvillanova in #1364
• Fix Llama model name in docs example by @SaiDunoyer in #1379
• fix typo in docstring in mcp_client.py by @grll in #1380
• Document use_structured_outputs_internally with version added by @albertvillanova in #1388
• Fix rendering of version added in docs by @albertvillanova in #1389
• Fix broken link in agentic RAG examples page by @vksx in #1363

Maintenance 🏗️

• Bump dev version: v1.17.0.dev0 by @albertvillanova in #1336
• Remove deprecated from_hf_api methods by @albertvillanova in #1351

New Contributors

• @FlorianVal made their first contribution in #1344
• @vksx made their first contribution in #1363
• @akseljoonas made their first contribution in #1346
• @SaiDunoyer made their first contribution in #1379

Full Changelog: v1.16.1...v1.17.0

v1.16.1

Bug Fixes 🐛

• Revert "Add timeout mechanism for code execution" by @albertvillanova in #1340

Full Changelog: v1.16.0...v1.16.1

v1.16.0

New features 🚀

• Tools
  • Add Bing search engine to WebSearchTool by @albertvillanova in #1313
• Local Python executor improvements
  • Support passing custom functions to local Python executor by @albertvillanova in #1302
  • Support executor_kwargs in LocalPythonExecutor for initialization customization by @albertvillanova in #1314
  • Add timeout mechanism for code execution by @albertvillanova in #1318
• CLI improvements
• Enable local web agents via api_base and api_key by @zbloss in #1333

Important changes ⚠️

• 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 #1322

Bug Fixes 🐛

• Streaming GradioUI fixes
  • Fix thought yield in GradioUI for streaming and non-streaming by @albertvillanova in #1311
  • Fix duplicate error message in GradioUI, test, and refactor by @albertvillanova in #1312
  • Fix duplicate plan display in GradioUI when streaming by @albertvillanova in #1317

Documentation 📚

• Improve docs about Agentic RAG by @albertvillanova in #1319
• Update Readme video with streaming example by @aymeric-roucher in #1306

Misc 🛠️

• Bump dev version: v1.16.0.dev0 by @albertvillanova in #1304
• Update minimum supported Gradio version to 5.14.0 to fix Sidebar-related AttributeError by @albertvillanova in #1309
• Fix deprecation of HfApiModel by @albertvillanova in #1315
• Fix CI DuckDuckGoSearchTool tests raising 202 Ratelimit by @albertvillanova in #1324
• Remove utils dir from Makefile check_dirs by @tobiasofsn in #1335

New Contributors

• @tobiasofsn made their first contribution in #1335
• @zbloss made their first contribution in #1333

Full Changelog: v1.15.0...v1.16.0

v1.15.0

New features 🚀

• Streaming support

  • Streaming model outputs by @aymeric-roucher in #1236
  • Working streaming Gradio chatbot outputs by @aymeric-roucher in #1246
  • Simplify validation of stream by @albertvillanova in #1242
  • Fix return type hint for generate_stream and rename CompletionDelta to ChatMessageStreamDelta by @albertvillanova in #1247
  • Remove dead code related to non-CompletionDelta streaming results by @albertvillanova in #1248

• New models

  • Add LiteLLM Router model by @NishantBaheti in #1096

• New tools

  • Implement WebSearchTool by @albertvillanova in #1300
  • Replace DuckDuckGoSearchTool with WebSearchTool in docs by @albertvillanova in #1303

New documentation 📚

• Add Installation docs by @albertvillanova in #1293
• Add docs about Using different models: Gemini models by @albertvillanova in #1297

Fixes and improvements 🔧

• Fix typo in pydoc by @darxriggs in #1222
• Change language from semantic to lexical for BM25 by @HeardACat in #1230
• Move MCPClient to root-level lib and manage optional dependencies by @njbrake in #1238
• Update README.md by @aymeric-roucher in #1233
• Set __name__ to "__main__" in local Python interpreter by @albertvillanova in #1234
• Remove unused argument from the MultiStepAgent documentation by @damnever in #1250
• Add billing for inference providers model by @merveenoyan in #1260
• Fix Loading GAIA dataset by @aymeric-roucher in #1266
• Fix remote executor send_tools to skip pip install on empty package list by @albertvillanova in #1273
• Remove smolagents installation from remote executors by @albertvillanova in #1274
• Allow VisitWebpageTool to run on remote executors without installing smolagents by @keetrap in #1288
• Refactor RemotePythonExecutor.send_tools to call install_packages by @albertvillanova in #1281
• Remove unnecessary dependencies from DockerExecutor image by @albertvillanova in #1272
• Support tools with union type return by @albertvillanova in #1294
• Fix WikipediaSearchTool in remote executors by @albertvillanova in #1283
• docs: fix typo by @luisegarduno in #1253
• Fix Custom Final Answer Tool for ToolCallingAgent by @HeardACat in #1255

Maintenance 🏗️

• Bump dev version: v1.15.0.dev0 by @albertvillanova in #1221
• CI: Simplify testing by running all tests in a single command by @albertvillanova in #1218
• Fix deprecated typing by @albertvillanova in #1249
• Update Open DeepResearch requirements by @albertvillanova in #1265
• Lower Pillow dependency minimum version to 10.0.1 by @albertvillanova in #1270
• Hotfix CI: pin mcp < 1.7.0 by @albertvillanova in #1285
• Remove duckduckgo-search as required dependency by @albertvillanova in #1271
• Unpin mcp < 1.7.0 by @albertvillanova in #1289
• Remove markdownify as required dependency by @albertvillanova in #1275

New Contributors

• @darxriggs made their first contribution in #1222
• @HeardACat made their first contribution in #1230
• @NishantBaheti made their first contribution in #1096
• @damnever made their first contribution in #1250
• @luisegarduno made their first contribution in #1253

Full Changelog: v1.14.0...v1.15.0

v1.14.0

New features 🚀

• SmolAgents MCP Client by @grll in #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 #1115
  • Add native support for Amazon Bedrock in the models module, enabling seamless interaction with Bedrock-hosted models.

Important changes 💥

• Rename HfApiModel to InferenceClientModel by @julien-c in #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 #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 🔧

• Unpin pillow < 11.2.0 by @albertvillanova in #1140
• Pass vLLM (VLLMModel) model client params as client_kwargs by @sergiopaniego in #1137
• Docs: Add Code Agents in agency table by @aymeric-roucher in #1144
• Replace ChatMessage.from_hf_api with from_dict with asdict by @albertvillanova in #1076
• Make to_messages methods compatible with qwen vl by @aymeric-roucher in #1148
• Fix get_json_schema for keyword in description by @albertvillanova in #1146
• Support images in stream_to_gradio by @aymeric-roucher in #1154
• Fix: adjust PlanningStep user message to avoid azure's content filter by @bpoppinga in #1151
• Refactor code agent system prompt by @aymeric-roucher in #1153
• Fix bug authors by @aymeric-roucher in #1163
• Fix memory step model output in ToolCallingAgent by @keetrap in #1156
• Parsing model outputs: allow whitespaces after the header of code blobs by @aymeric-roucher in #1172
• Fix: Add default value to optional argument by @rca-umb in #1155
• Add more detail to code execution documentation by @maeste in #983
• Make TextInspectorTool serializable by @albertvillanova in #1176
• Clarify tool usage in agent step tutorial by @albertvillanova in #1175
• Provide ability to override Agent logger by @sysradium in #1166
• Pin mcpadapt >= 0.0.19 to include security fix by @albertvillanova in #1178
• Fix tool validation with multiple assignments by @albertvillanova in #1177
• Fix typo in the memory.mdx by @alpsencer in #1187
• feat: add custom Docker image support and enhanced configuration for DockerExecutor by @kingdomad in #1057
• Add api_key argument in HfApiModel by @aymeric-roucher in #947
• Implement Tool.from_dict by @albertvillanova in #1006
• Implement Agent.from_dict by @albertvillanova in #1007
• Call recursively Agent.to_dict for managed agents by @albertvillanova in #1008
• Support Literal type annotations in @tool for defining enums by @stephanlensky in #1195
• Support class docstrings in LocalPythonExecutor by @albertvillanova in #1192
• Reference Inference Providers more clearly by @julien-c in #1197
• Support annotated assignment in LocalPythonExecutor by @albertvillanova in #1193
• Fix evaluate_boolop for non-boolean return type by @ZeusFSX in #1191
• Make MultiStepAgent and abstract class by @sysradium in #1204
• fix: creating tool should not affect function that is being wrapped by @njbrake in #1203
• Remove header comments from MDX docs for proper rendering by @albertvillanova in #1212
• Fix CI test by updating LiteLLM error message by @albertvillanova in #1214
• Fix typo in evaluate_ast() docs: static_tools -> custom_tools by @JungeAlexander in #1217
• Make OpenDeepResearch tools serializable by @albertvillanova in #1206
• Add Security Policy by @albertvillanova in #1208
• Improve docs for stream run by @albertvillanova in #1215
• Skip unsupported stop parameter for reasoning models by @albertvillanova in #1211

New Contributors

• @leandrodamascena made their first contribution in #1115
• @bpoppinga made their first contribution in #1151
• @rca-umb made their first contribution in #1155
• @alpsencer made their first contribution in #1187
• @stephanlensky made their first contribution in #1195
• @ZeusFSX made their first contribution in #1191
• @JungeAlexander made their first contribution in #1217

Full Changelog: v1.13.0...v1.14.0

v1.13.0

Important changes 🚀

• Gradio UI now logs images observed by your agent: log observations images by @aymeric-roucher in #1113
• allow users to retrieve original Gradio app by @sysradium in #811
• Re-authorize submodule imports if top was allowed by @aymeric-roucher in #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 #1035
• Fix telemetry AttributeError: module 'smolagents' has no attribute 'ApiModel' by @IlyaGusev in #1040
• Hotfix CI: pin transformers < 4.50.0 by @albertvillanova in #1044
• Align name of tool source by @albertvillanova in #1042
• Update type hints and docstrings of Agent/Tool .save with Path type by @albertvillanova in #1036
• Fix PythonInterperterTool description of code input with auth imports by @benediktstroebl in #1053
• Fix typos in Code Agent prompt by @andysalerno in #1063
• Validate agent name by @Misakar-0v0 in #768
• fix(MLXModel): define var to avoid UnboundLocalError by @SauravMaheshkar in #1059
• Minor fix code block rendering in Open DeepResearch README by @keetrap in #1055
• Add WikipediaSearchTool to default tools by @touseefahmed96 in #514
• Improve MLXModel call by @albertvillanova in #1067
• Support transformers 4.50.0 by @albertvillanova in #1070
• Improve Tool.push_to_hub to avoid saving to local filesystem by @albertvillanova in #1051
• Fix Tool.to_dict for nullable input parameter by @albertvillanova in #1038
• Support tool Gradio app with boolean input type by @albertvillanova in #1065
• Support running CI on push to ci-* branch by @albertvillanova in #1047
• Set Agent/Tool requirements as a sorted list by @albertvillanova in #1048
• Support passing raw to ChatMessage.from_dict by @albertvillanova in #1010
• Distinguish between AgentToolCallError and AgentToolExecutionError by @sysradium in #1060
• Add detail on potential threats to system by @aymeric-roucher in #1077
• Refactor Tool.save by @albertvillanova in #1075
• Fixing typos in toolcallingagent prompt by @Y-artios in #1082
• Refactor step yielding during streaming run to log PlanningSteps by @aymeric-roucher in #1083
• Make pandas a test extra dependency only by @albertvillanova in #1081
• Add security disclaimer to MCP docs by @albertvillanova in #1090
• Fix get_json_schema for docstrings with type annotations by @albertvillanova in #1088
• Remove transformers/huggingface-hub to check torch availability by @albertvillanova in #1022
• Improve ApiModel and subclasses instantiation by @albertvillanova in #1095
• Add trust_remote_code to ToolCollection.from_mcp by @albertvillanova in #1091
• Add user message after plan by @aymeric-roucher in #1104
• Add agent interruption by @aymeric-roucher in #1112
• Validate tool name as Python identifier by @albertvillanova in #1041
• Increase request timeout for tests that time out by @albertvillanova in #972
• Update README.md by @aymeric-roucher in #1125
• Change: expanded error exception for incorrect tool parameters by @cgthayer in #1099
• Hotfix CI AttributeError: 'PngImageFile' object has no attribute 'filename' by @albertvillanova in #1128
• Replace AgentGenerationError with AgentParsingError for ToolCallingAgent by @aymeric-roucher in #1131
• Fix planning_interval=1 bug in MultiStepAgent by @RolandJAAI in #1118
• Set main as the no-argument CLI entry point by @albertvillanova in #1126

New Contributors

• @benediktstroebl made their first contribution in #1053
• @andysalerno made their first contribution in #1063
• @Misakar-0v0 made their first contribution in #768
• @SauravMaheshkar made their first contribution in #1059
• @Y-artios made their first contribution in #1082
• @cgthayer made their first contribution in #1099

Full Changelog: v1.12.0...v1.13.0

v1.12.0

What's Changed

• Bump dev version: v1.12.0.dev0 by @albertvillanova in #982
• Fix json parsing by @sysradium in #987
• Fix Agent.from_hub for older smolagents exports by @albertvillanova in #989
• Regression fix: ChatMessageToolCall id must be a string by @sysradium in #990
• Always parse tool calls from APIs by @aymeric-roucher in #991
• Sort additional authorized imports by @keetrap in #998
• Fix custom prompt templates in Agent.from_hub by @albertvillanova in #1001
• Always parse arguments from API tool calls by @aymeric-roucher in #1000
• Fix TransformersModel: torch_dtype and trust_remote_code not passed to VLM by @Louis-Gv in #1012
• Fix double space in toolcalling_agent.yaml by @louisbrulenaudet in #1021
• Update images type and docstring by @keetrap in #1018
• Support JSON string arguments in tool calls by @albertvillanova in #1003
• Pass trust_remote_code to transformers tokenizer/processor by @albertvillanova in #1013
• Fix code using pillow required dependency by @albertvillanova in #1019
• Fix tests of Agent.save and Tool.save by @albertvillanova in #1029
• Remove dependency on transformers get_tests_dir by @albertvillanova in #1026
• Reduce model calls in planning_step from 2 to 1 by @aymeric-roucher in #985
• Bugfix: adding <end_code> to prevent imaginary tool calls in CodeAgent by @IlyaGusev in #918
• Use pytest-datadir for shared data files by @albertvillanova in #1031
• Refactor Agent.to_dict by @albertvillanova in #1004
• Add support for MCP SSE servers by @grll in #931

New Contributors

• @Louis-Gv made their first contribution in #1012
• @louisbrulenaudet made their first contribution in #1021

Full Changelog: v1.11.0...v1.12.0