NewYour coding agent can read the release notes before it upgrades.Set up the MCP server →
pub.dev
Shared OpenAI Chat Completions base for AI SDK Dart providers. Implements tool calling, multimodal content, structured output, and SSE streaming once, behind a small per-provider config.
Last release 3 days ago
04 Oct 2026
Ships on a steady schedule
a new release about every 2 months
Nearly every release is documented
notes for 3 of 3 stable releases
Nothing withdrawn
no release was ever pulled
3 months old
3 releases · first in 2026
One column per month.
LanguageModelV4AbortSignal remains as a deprecated alias of AbortSignal for the whole 3.x line.
AI SDK Dart 3.0 is a coordinated major release across 17 packages, including four new packages.
AbortSignal reaches the live HTTP request in every built-in provider. A stream that ends early fails with a typed truncation error.responseMessages now contains only the messages that the current call generated. embedMany separates batch size from concurrency and validates the rows it gets back.ai_sdk_conversation: versioned, immutable conversation snapshots with a JSON persistence codec.ai_sdk_remote: Vercel AI UI-message stream transport for trusted backends.ai_sdk_json_schema: optional local JSON Schema validation (Draft 7 and 2020-12) for structured output.ai_sdk_telemetry: bounded OTLP/HTTP JSON metric export.Upgrade every ai_sdk package together. Custom embedding models must declare maxEmbeddingsPerCall and supportsParallelCalls. A decoder alone no longer means JSON Schema validation: use Schema.decoderOnly or ai_sdk_json_schema. LanguageModelV4AbortSignal remains as a deprecated alias of AbortSignal for the whole 3.x line.
Read the 2.x to 3.x migration guide before you upgrade. Each package's CHANGELOG lists its own changes.
All 17 packages are published on pub.dev. ai_sdk_realtime stays unpublished until it completes qualification.
Full Changelog: 2.0.0...3.0.0
AI SDK Dart 2.0 is a coordinated major release across all 13 packages.
AI SDK Dart 2.0 is a coordinated major release across all 13 packages.
Upgrade every ai_sdk package together. The obsolete V3 language-model contract, experimental API names, video surface, and old provider call fields were removed without compatibility shims.
Read the 2.0 migration guide before upgrading.
All 13 packages are published on pub.dev.
Full Changelog: 1.2.0...2.0.0
doStream) and non-streaming (doGenerate) paths
now surface model reasoning/thinking. Streaming delta reasoning is emitted
as StreamPartReasoningDelta, and non-streaming message reasoning as a
LanguageModelV4ReasoningPart, so streamText(...).fullStream /
result.reasoning and generateText(...).reasoning are populated when the
provider returns it. Previously reasoning was silently dropped.OpenAICompatibleConfig.reasoningKeys — the response field names
checked (in order) for reasoning text. Defaults to
['reasoning_content', 'reasoning', 'thinking'], covering DeepSeek
(reasoning_content), OpenRouter (reasoning), and thinking-style hosts;
set to const [] to disable extraction. Benefits every provider built on this
base (OpenAI, Azure, Groq, Mistral, …).Coordinated 1.2.0 release across all 13 packages — now live on pub.dev. A v6 hardening pass that makes the AI SDK Dart port more complete, honest, and
Coordinated 1.2.0 release across all 13 packages — now live on pub.dev. A v6 hardening pass that makes the AI SDK Dart port more complete, honest, and tested.
Non-2xx provider responses now throw a typed AiApiCallError carrying the provider's message, type, code, statusCode, raw body, and an isRetryable flag — consistently across every provider. The AiSdkError hierarchy moved into ai_sdk_provider (re-exported by ai_sdk_dart, so the public surface is unchanged).
ai_sdk_openai_compatible base (new package)A shared OpenAI Chat Completions implementation now powers OpenAI / Azure / Groq / Mistral. All 8 providers support tool use + multimodal image input; Cohere & Ollama gained them natively.
Real Server-Sent-Events streaming transport (SseClientTransport) plus a plain request/response HttpClientTransport; web-safe — dart:io is isolated behind conditional imports, so the client runs on Flutter web.
19 prebuilt, themeable Material widgets plus hardened controllers (ObjectStreamController.submit, ChatController.isStreaming, streaming-error surfacing). A restrained Material 3 redesign — bubbleless assistant transcript, an AiMotion motion/haptics vocabulary, and a first-class reduced-motion path. Fixed a PressableScale crash-on-dispose under reduced motion.
onAbort now interrupts an in-flight streamText; stopWhen can extend the tool loop past maxSteps. Removed the unusable generateVideo / VideoModelV1 (no backing provider). OpenAI image generation no longer sends response_format to gpt-image-1.
1,057 tests (924 Dart + 133 Flutter), 99.9% line coverage (11 of 12 packages at 100%), enforced by a CI coverage gate.
ai_sdk_provider · ai_sdk_dart · ai_sdk_openai_compatible · ai_sdk_openai · ai_sdk_anthropic · ai_sdk_google · ai_sdk_azure · ai_sdk_cohere · ai_sdk_groq · ai_sdk_mistral · ai_sdk_ollama · ai_sdk_mcp · ai_sdk_flutter_ui
See each package's CHANGELOG.md for full details.
Full Changelog: 1.1.0...1.2.0
First release (versioned 1.2.0 to align with the AI SDK Dart monorepo).
OpenAICompatibleChatLanguageModel — a full LanguageModelV3 implementation
of the OpenAI Chat Completions wire format: multimodal content parts (text +
image + audio + file), tools / tool_choice (with strict),
response_format JSON-schema, SSE streaming (text deltas + index-based
tool-call delta state machine + finish + usage), non-streaming tool-call
parsing, and finish-reason mapping.apiErrorFromDioException(...) — a shared Dio→AiApiCallError bridge (delegating to
AiApiCallError.fromResponse) so non-2xx responses surface as a typed AiApiCallError on both
the non-streaming and stream-setup paths. Imported by the OpenAI / Azure / Groq / Mistral
providers.OpenAICompatibleConfig — a small config object parameterizing the
per-provider quirks: auth header scheme, base URL, query parameters
(e.g. Azure api-version), body field name overrides (seed vs
random_seed, max_tokens vs max_completion_tokens), feature flags
(supportsTools, supportsMultimodal, supportsResponseFormatJsonSchema),
an extraBody hook for provider-specific fields, and a Dio factory.Your coding agent can read these notes before it upgrades. Set up the MCP server →