NewYour coding agent can read the release notes before it upgrades.Set up the MCP server →
npm · #2521 most downloaded on npm
Last release today
05 Oct 2026
Ships on a steady schedule
a new release about every 8 days
Some releases are documented
notes for 31 of the last 60 stable releases
Nothing withdrawn
no release was ever pulled
1 years old
761 releases · first in 2025
One column per month.
@ai-sdk/provider-utils@4.0.0-beta.41
b1624f0: feat (provider/gateway): add trinity-mini model id
@ai-sdk/provider-utils@4.0.0-beta.40
@ai-sdk/provider-utils@4.0.0-beta.39
@ai-sdk/provider-utils@4.0.0-beta.38
cdd0bc2: feat (provider/gateway): add intellect-3 model id
457318b: chore(provider,ai): switch to SharedV3Warning and unified warnings
8d9e8ad: chore(provider): remove generics from EmbeddingModelV3
8d9e8ad: chore(provider): remove generics from EmbeddingModelV3
Before
model.textEmbeddingModel("my-model-id");
After
model.embeddingModel("my-model-id");
Updated dependencies [8d9e8ad]
@ai-sdk/provider-utils@4.0.0-beta.35
e8694af: feat(provider/gateway): Server-side image request splitting
@ai-sdk/provider-utils@4.0.0-beta.34
5dd4c6a: fix(provider/gateway): Fix Gateway image model provider options not passing through
1425df5: feat(provider/gateway): Add Imagen 4 Ultra Generate to model string autocomplete list
bca7e61: feat(provider/gateway): Change default maxImagesPerCall per-provider
2d166e4: feat(provider/gateway): add support for image models
cc5170d: feat(provider/gateway): update gateway model string autocomplete
5f66123: chore(provider/gateway): Update gateway language model settings
3782645: bump @vercel/oidc to 3.0.5
@vercel/oidc to 3.0.5@ai-sdk/provider-utils@4.0.0-beta.33
@ai-sdk/provider-utils@4.0.0-beta.32
@ai-sdk/provider-utils@4.0.0-beta.31
@ai-sdk/provider-utils@4.0.0-beta.30
@ai-sdk/provider-utils@4.0.0-beta.29
96322b7: feat(provider/gateway): Add GPT OSS Safeguard 20B to Gateway model string autocomplete
@ai-sdk/provider-utils@4.0.0-beta.28
@ai-sdk/provider-utils@4.0.0-beta.27
7d73922: feat(provider/gateway): Add MiniMax M2 to Gateway autocomplete
@ai-sdk/provider-utils@4.0.0-beta.26
@ai-sdk/provider-utils@4.0.0-beta.25
0c3b58b: fix(provider): add specificationVersion to ProviderV3
@ai-sdk/provider-utils@4.0.0-beta.23
@ai-sdk/provider-utils@4.0.0-beta.22
@ai-sdk/provider-utils@4.0.0-beta.21
2b6a848: feat (provider/gateway): add models provider option for model routing
@ai-sdk/provider-utils@4.0.0-beta.20
@ai-sdk/provider-utils@4.0.0-beta.19
2f8b0c8: fix(gateway): bump @vercel/oidc to latest
@vercel/oidc to latest1890317: feat (provider/gateway): improve auth error messages
@ai-sdk/provider-utils@4.0.0-beta.18
@ai-sdk/provider-utils@4.0.0-beta.17
0a2ff8a: feat (provider/gateway): add user and tags provider options
@ai-sdk/provider-utils@4.0.0-beta.16
f0b2157: fix: revert zod import change
@ai-sdk/provider-utils@4.0.0-beta.14
@ai-sdk/provider-utils@4.0.0-beta.13
@ai-sdk/provider-utils@4.0.0-beta.12
0e29b8b: chore(provider/gateway): lazy schema loading
acc14d8: feat (provider/gateway): add 'only' to provider options
95f65c2: chore: use import \* from zod/v4
7b1b1b1: fix(provider/gateway): add "react-native" as export condition for browser behavior
7b1b1b1: fix(provider/gateway): add "react-native" as export condition for browser behavior
This avoids the use of native Node APIs in bundles created for React Native / Expo apps
@ai-sdk/provider-utils@4.0.0-beta.10
@ai-sdk/provider-utils@3.1.0-beta.9
@ai-sdk/provider-utils@3.1.0-beta.8
1d8ea2c: feat(provider/gateway): Add GPT-5 pro to Gateway model string autocomplete
@ai-sdk/provider-utils@3.1.0-beta.7
ef62178: feat(gateway): oidc refresh with @vercel/oidc
@vercel/oidca90dca6: feat(provider/gateway): Add zAI GLM 4.6 to Gateway language model settings
0adc679: feat(provider): shared spec v3
e6bfe91: feat(provider/gateway): Update DeepSeek model string autocomplete
636e614: feat(provider/gateway): Add DeepSeek V3.2 Exp to Gateway language model settings
Your coding agent can read these notes before it upgrades. Set up the MCP server →