Rivet IDE v1.7.6

Bug Fixes

• Fix regression with rendering objects as strings causing crashes or hanging UI

Rivet Libraries v1.14.1

Bug Fixes & Improvements

• Fix Destructure Node wrapping single-value results in arrays. If you do $.path and it only matches one thing, you will get "result" out instead of ["result"]
• Validate HTTP Call Node URLs before calling them
• Improve Object Node type inference - if the type being returned is an object, it will correctly return object[] instead of object to improve type coercion in later nodes.
• Improve type coercion for type: 'object' values. This is now treated more or less the same as an any (the actual type is inferred by the value's type). This improves coercion to string for arrays.
• Fix isSplitSequential being serialized to project file.

Rivet IDE v1.7.5

Bug Fixes & Improvements

• Persist last debugger connection URI
• Fix flickering of nodes while dragging of them
• Fix fsPath error when opening new project
• Fix hotkeys on Windows
• Fix LM Studio auto-configure path
• Limit Object Node body rendering length so that it doesn't expand infinitely
• Fix errors when running Rivet in a browser
• Fix constant memory usage increase on Windows
• Fix high CPU usage when Rivet is idle
• Fix console error when dragging wire around
• Fix Destructure Node wrapping its outputs in an array, when there is only one match for the output.
• Validate URLs for the HTTP Call Node
• Object Node is now more intelligent about its output data type. You can use arrays in an Object Node and they will be interpreted correctly when coercing in subsequent nodes.
• Fix coercion from object to string, when the object is really an array. Object and any function more or less the same now (by analyzing the contents to see what type it is)
• Fix saving "split sequential" toggle

Rivet Libraries v1.14.0

New Features

• Google Gemini Plugin.

Bug Fixes

• Packages updated.
• AssemblyAI plugin and package updated.
• Fix concurrent chat node with local LLM (#249)
• Fix creating objects with type and value properties.
• Fix abort/subscribe memory leak.

Rivet IDE v1.7.4

New Features

• Added Google plugin! Use Google Gemini.

Changes & Fixes

• Plugin UI: Updated image thumbnails for Autoevals, Python, and FS plugins.
• Fix JavaScript example in Code Node.
• Fix input/output names in Code Node was crashing render.
• Packages updated.
• AssemblyAI plugin and package updated.
• Fix concurrent chat node with local LLM (#249)
• Fix creating objects with type and value properties.
• Fix abort/subscribe memory leak.

Rivet Libraries v1.13.4

Bug Fixes

• Fix circular import with ESM importing (might be esbuild specific)

Rivet Libraries v1.13.3

Fixes

• Fix base64 encoding when executing in Node.js, which fixes GPT-4 Vision being used there

Rivet IDE v1.7.3

New Features

• Added Ollama plugin! Use Ollama to run local LLM models.

Changes & Fixes

• OpenAI Plugin: Allow more values (better coercion) to be passed into the messages input of the Run Thread node.
• OpenAI Plugin: Fix data type for functions input of Run Thread node
• OpenAI Plugin: Include message input to "On Message Creation" subgraph like the helper text says
• OpenAI Plugin: Fix error message when a function call isn't mapped in the handlers
• Added helper tooltip to all input port toggle switches
• Fixed base64 encoding images for GPT-4 Vision when using Node executor

Rivet Libraries v1.13.2

Tweaks & Fixes

• Fixed NodeDatasetProvider not preserving options such as save when it does not find a data file.
• Added requireFile option to NodeDatasetProvider, so that it can error when it does not find a dataset file.
• Better coercion of messages input for Run Thread node.
• Fix a Run Thread node error message.
• Correct functions input data type for Run Thread node.
• Run Thread passes a message input for the "on message creation subgraph" like the docs say

Rivet Libraries v1.13.1

Tweaks & Bug Fixes

• Fixed request token count being 0 when number of choices is unset (1)
• Account for GPT functions in token counting by approximating what OpenAI does to calculate their token counts.