v0.4.6 (2025-07-23)

What's Changed

• Enable URL and binary PDF for Mistral by @pintaf in #2267
• Fix pydantic-evals panel rendering with evaluators by @dmontagu in #2274
• Fix mp3 handling by @dmontagu in #2279
• Handle built-in tool errors better in tool registration by @fswair in #2252
• fix: use FileUrl.format to find the extension by @Kludex in #2280
• Correct code snippet for native output by @minhduc0711 in #2271
• chore: simplify output function call with model retry by @bitnahian in #2273
• Reduce duplication between StreamedRunResult and AgentStream by @DouweM in #2275
• chore: add CLAUDE.md by @Kludex in #2281
• Speed up function _estimate_string_tokens by @misrasaurabh1 in #2156

New Contributors

• @minhduc0711 made their first contribution in #2271
• @pintaf made their first contribution in #2267

Full Changelog: v0.4.5...v0.4.6

v0.4.5 (2025-07-22)

What's Changed

• Add async with self in agent_to_a2a by @Kludex in #2266
• Fix include_content not working as expected by @adtyavrdhn in #2206
• Support streamable HTTP in mcp-run-python by @Kigstn in #2230
• change format_as_xml defaults by @samuelcolvin in #2228
• Fix LLMJudge input handling to preserve BinaryContent as separate message part instead of stringifying by @adtyavrdhn in #2173
• validate OpenAI responses by @samuelcolvin in #2226
• Remove duplicate field on GeminiModelSettings by @strawgate in #2269
• Fix AG-UI shared state example by @stevenh in #2272

New Contributors

• @Kigstn made their first contribution in #2230
• @strawgate made their first contribution in #2269

Full Changelog: v0.4.4...v0.4.5

v0.4.4 (2025-07-18)

What's Changed

• Add Toolsets and Deferred Tools by @DouweM in #2024
• Support AG-UI protocol for frontend-agent communication by @stevenh in #2223
• Add identifier field to BinaryContent class by @GDaamn in #2231
• Add OpenAI models o1-pro, o3-pro, o3-deep-research, computer-use by @zachmayer in #2234
• Add grok-4 and groq kimi-k2 models by @zachmayer in #2235
• Remove old Google models by @DouweM in #2220
• Change clai default model to gpt-4.1 by @samuelcolvin in #2227
• Fix VertexAI Empty Model Parts Error by @caesarnine in #2203
• Remove incorrect tool call id from OpenAI tool call delta by @naveen-corpusant in #2210
• Speed up method AgentRunResult._set_output_tool_return by 18798% by @misrasaurabh1 in #2196
• Speed up method Usage.opentelemetry_attributes by 85% by @misrasaurabh1 in #2198
• Nicer errors under the capture_run_messages context by @alonme in #2219

New Contributors

• @naveen-corpusant made their first contribution in #2210
• @caesarnine made their first contribution in #2203
• @GSemikozov made their first contribution in #2193
• @alonme made their first contribution in #2219
• @GDaamn made their first contribution in #2231

Full Changelog: v0.4.3...v0.4.4

v0.4.3 (2025-07-16)

What's Changed

• Include content.1571 by @adtyavrdhn in #2180
• duckduckgo-search is renamed to ddgs by @basnijholt in #2172
• Add base64 encoding to tool_return_ta by @adtyavrdhn in #2186
• Bugfix: avoid race condition when refreshing google token by @JadHADDAD92 in #2100
• feat: Add output function tracing by @bitnahian in #2191
• Add Hugging Face as a provider by @hanouticelina in #1911

New Contributors

• @mmabrouk made their first contribution in #2192
• @basnijholt made their first contribution in #2172
• @JadHADDAD92 made their first contribution in #2100
• @bitnahian made their first contribution in #2191
• @hanouticelina made their first contribution in #1911

Full Changelog: v0.4.2...v0.4.3

v0.4.2 (2025-07-10)

What's Changed

• Let model settings be passed to model classes by @svilupp in #2136
• Add StructuredDict for structured outputs with custom JSON schema by @fswair in #2157
• Handle DeepSeek reasoning_content in streamed responses by @tarruda in #2174
• Drop FastA2A from PydanticAI repository by @Kludex in #2171
• Fix type annotations for Agent.iter() by @erosennin in #2168
• Fix chat-app example doc - python code appear twice by @itayB in #2169
• Speed up function _ensure_decodeable by 634% by @misrasaurabh1 in #2155

New Contributors

• @misrasaurabh1 made their first contribution in #2155
• @erosennin made their first contribution in #2168
• @itayB made their first contribution in #2169
• @tarruda made their first contribution in #2174

Full Changelog: v0.4.1...v0.4.2

v0.4.1 (2025-07-08)

What's Changed

• Support evaluating sync tasks by @dmontagu in #2150
• Upgrade a2a to spec v0.2.5 by @physicsrob in #2144
• Drop FastA2A as PydanticAI dependency by @Kludex in #2164

New Contributors

• @physicsrob made their first contribution in #2144

Full Changelog: v0.4.0...v0.4.1

v0.4.0 (2025-07-08)

What's Changed

• BREAKING CHANGE: Make EvaluationReport and ReportCase into generic dataclasses by @dmontagu in #1799
• Make ToolDefinition.description optional and fix Bedrock description handling by @dmontagu in #1507
• Add all audio types supported by Gemini to AudioUrl by @ChenghaoMou in #2151
• Improve number_to_datetime performance by building TypeAdapter only once by @DouweM in #2153
• Retain defaults in non-strict openai schemas by @dmontagu in #1519

Full Changelog: v0.3.7...v0.4.0

v0.3.7 (2025-07-07)

What's Changed

• Make AgentStream.stream_output (available inside agent.iter) stream validated output data instead of raising validation errors by @DouweM in #2134
• Add model_request_stream_sync to direct API by @hewliyang in #2116
• Add GitHub Models provider by @sgoedecke in #2114
• Added support for google specific arguments for video analysis by @Sumered in #2110
• Implemented a convenient way to use ACI.dev Tools in PydanticAI by @Kamal-Moha in #2093
• Fix list rendering in documentation by @Viicos in #2145
• Raise consistent deprecation warnings by @Viicos in #2148
• Move docstring warning of model settings as a comment by @Viicos in #2146

New Contributors

• @hewliyang made their first contribution in #2116
• @sgoedecke made their first contribution in #2114
• @Kamal-Moha made their first contribution in #2093
• @Sumered made their first contribution in #2110

Full Changelog: v0.3.6...v0.3.7

v0.3.6 (2025-07-04)

What's Changed

• Deprecate {FunctionToolCallEvent,FunctionToolResultEvent}.call_id in favor of tool_call_id by @proever in #2028
• Indicate to the model that a RetryPromptPart not tied to a tool call contains validation feedback rather than a user message by @hovi in #2008
• Update starlette subdomain in docs by @alDuncanson in #2099
• Update client.md - Typo by @kauabh in #2105
• Add support for predicted outputs in OpenAIModelSettings by @webholics in #2106
• Record tool response in tool run span by @alexmojaki in #2109
• Use contextvars for agent overriding, rather than a local attribute by @dmontagu in #2118
• Fix model parameters not being customized in fallback model request stream by @almeidaalajoel in #2120
• simplify weather example by @samuelcolvin in #2129

New Contributors

• @alDuncanson made their first contribution in #2099
• @webholics made their first contribution in #2106
• @almeidaalajoel made their first contribution in #2120

Full Changelog: v0.3.5...v0.3.6

v0.3.5 (2025-06-30)

What's Changed

• Add progress bar on evaluate by @davide-andreoli in #1871
• Fix deprecation warning under Pydantic 2.11 by @medaminezghal in #2076
• fix: async fixtures in conftest.py by @stevenh in #2068
• fix: docs examples python version in tests by @stevenh in #2069
• Set 'us-central1' by default on GoogleProvider by @Kludex in #2031
• Move ThinkingPart to preceed TextPart in OpenAIResponsesModel by @Kludex in #2043
• Fix deprecated kwargs validation to prevent silent failures by @svilupp in #2047
• Support strict mode in NativeOutput by @severinh in #2084
• Let tools return ToolReturn to pass additional content to model, or attach metadata that's not passed to the model by @Wh1isper in #2060
• Add ability to include snippets in docs with inline-named sections for fragments and highlighting by @dmontagu in #2088
• Add Slack Lead Qualifier example by @DouweM in #2079

New Contributors

• @medaminezghal made their first contribution in #2076
• @svilupp made their first contribution in #2047
• @severinh made their first contribution in #2084

Full Changelog: v0.3.4...v0.3.5