npm / @ai-sdk/mistral
@ai-sdk/mistral
What changed in @ai-sdk/mistral 4, from its official release notes, upgrade guides and advisories. Sources checked 2026-10-18.
1 security4 new APIs and options2 behavior and defaults10 fixes and other
See which of these apply to your code, ranked, in your agent:
npx @unstale/cli init@ai-sdk/mistral 417 changes
Security 1
@ai-sdk/provider-utils has an Uncontrolled Resource Consumption issue
New APIs and options 4
Adds a conversations API to @ai-sdk/mistral.
Adds a web search tool to @ai-sdk/mistral.
The portable reasoning levels now include max, widening LanguageModelV4CallOptions['reasoning'] and ReasoningLevel. Consumers with exhaustive switches or mappings must handle it.
Experimental streaming lifecycle callbacks were added for speech generation and transcription.
Behavior and default changes 2
Top-level embedding dimensions are mapped to provider settings.
Telemetry was added to speech generation and transcription, with provider usage propagation.
Fixes and other changes 10
Reasoning configuration is forwarded for unknown model IDs instead of being silently stripped; known models without reasoning-effort support retain their previous handling.
Reasoning effort is now enabled for Mistral Large 4 models.
JSON schemas are preserved when structured outputs are disabled.
Provider error finish reasons are mapped to error.
The User-Agent header is now standards-compliant.
Fixes streamed tool calls with missing, blank, or repeated IDs.
Updated Mistral model IDs.
Updated the Mistral reasoning configuration.
Raw usage metadata now preserves complete Mistral chat usage objects.
Mistral now supports incremental streaming tool calls.