NewYour coding agent can read the release notes before it upgrades.Set up the MCP server →
npm · #2684 most downloaded on npm
AI SDK by Vercel - build apps like ChatGPT, Claude, Gemini, and more with a single interface for any model using the Vercel AI Gateway or go direct to OpenAI, Anthropic, Google, or any other model provider.
Last release today
06 Oct 2026
Ships on a steady schedule
a new release about every 8 days
Some releases are documented
notes for 27 of the last 60 stable releases
Nothing withdrawn
no release was ever pulled
13 years old
1572 releases · first in 2014
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
One column per quarter.
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
76024fc: fix(ai): fix static tool call and result detection when dynamic is undefined
### Patch Changes - Updated dependencies [f49f924] - @ai-sdk/gateway@1.0.23
0294b58: feat(ai): set ai, @ai-sdk/provider-utils, and runtime in user-agent header
ai, @ai-sdk/provider-utils, and runtime in user-agent headerde5c066: fix(ai): forwarded providerExecuted flag in validateUIMessages
cd91e4b: fix(ai): use correct type for reasoning outputs
### Patch Changes - Updated dependencies [4ee3719] - @ai-sdk/gateway@1.0.21
a0a725f: feat (ai): export createGateway
### Patch Changes - Updated dependencies [350a328] - @ai-sdk/gateway@1.0.20
d6785d7: feat (ai): add tool and agent helpers
ccc2ded: feat (ai): export gateway provider
### Patch Changes - 99c946a: export missing type
### Patch Changes - Updated dependencies - @ai-sdk/gateway@1.0.19
### Patch Changes - Updated dependencies [5d59a8c] - @ai-sdk/gateway@1.0.18
### Patch Changes - Updated dependencies [b6005cd] - @ai-sdk/gateway@1.0.17
### Patch Changes - Updated dependencies [99964ed] - @ai-sdk/provider-utils@3.0.8 - @ai-sdk/gateway@1.0.16
7fcc6be: feat(ai): throw InvalidArgumentError when messages is not provided
e0e9449: feat(ui): sent isAbort, isDisconnect, isError in useChat onFinish callback
4b81e7d: fix(ai): remove vitest dependency from test exports
ca40fac: feat(ai): support custom download functions (experimental)
33cf848: feat(ai): pass messages to useChat({ onFinish })
useChat({ onFinish })ca65923: fix(ai): remove use of expect() from production code
expect() from production codef8f3682: fix: call onFinish when stream is cancelled in toUIMessageStream
f8f3682: fix: call onFinish when stream is cancelled in toUIMessageStream
Previously, onFinish was only called on normal stream completion. Now it's also called when the reader is cancelled (e.g., browser close, navigation), ensuring partial messages are persisted.
Updated dependencies
5099b3d: fix(ai): make chat.addToolResult() compatible with dynamic tool calls
chat.addToolResult() compatible with dynamic tool calls### Patch Changes - Updated dependencies - @ai-sdk/gateway@1.0.11
581abea: fix(ai): call abort callback when stream is aborted during tool execution
8a87693: fix(ai) Make sure warnings promise in streamObject is resolved and properly collects and passes warnings
8da6e9c: fix(ai): use parsed tool input if possible when validation fails
### Patch Changes - Updated dependencies [8b96f99] - @ai-sdk/gateway@1.0.9
Updated dependencies [ d3bcad9 ]
### Patch Changes - Updated dependencies [68751f9] - @ai-sdk/provider-utils@3.0.4 - @ai-sdk/gateway@1.0.8
30f32bd : feat(xai): add grok-imagine-image-2.0 image model id
auto image quality option1.5k image resolution provider optionvadThreshold, and opus audio formatUpdated dependencies [ 8c65988 ]
ede5b89 : chore: migrate package builds from tsup to tsdown
a7b2e66: Added providerOptions to agent stream and generate calls
9bed210: ### extractReasoningMiddleware(): delay sending text-start chunk to prevent rendering final text before reasoning
When wrapping a text stream in extractReasoningMiddleware(), delay queing the text-start chunk until either reasoning-start chunk was queued or the first text-delta chunk is about to be queued, whichever comes first.
https://github.com/vercel/ai/pull/8036
c2511c1 : fix: use standards-compliant User-Agent header
525efc5 : feat(provider): advertise image model file and mask input support
525efc5: feat(provider): advertise image model file and mask input support
Use confirmed model IDs for capability declarations so unrecognized model names
remain unknown. Include Together AI FLUX.2 Pro and Flex single-image editing, and
allow asynchronous capability lookups and middleware overrides to resolve to
unknown.
Advertise QuiverAI Arrow 2 and Arrow 2 Telos file-input support, and mark Together
AI Gemini image inputs unsupported by the current single-image request mapping.
Updated dependencies [e3605f6]
Updated dependencies [525efc5]
Your coding agent can read these notes before it upgrades. Set up the MCP server →