Ollama v0.40.17 Adds Local Support for Inkling 2 Mini
Ollama v0.40.17 adds local support for Thinking Machines’ Inkling 2 Mini, a multimodal model that handles text and images and supports tool calls and thinking. Users can launch it with the command “ollama run inkling-2-mini.” The release supports Apple silicon only, while Ollama says PC support will follow in an upcoming release. Ollama also says the model can fit on a Mac with about 20GB of memory.
The release provides a documented way to run Inkling 2 Mini locally on Apple silicon, with multimodal and tool-calling capabilities; PC support is not included in this release.
Inkling 2 Mini
Inkling 2 Mini from Thinking Machines is a 26B parameter model with 5.4B active.
Its looped layers give it more computation per token than its size suggests, and it still fits on a Mac at about 20GB. It handles text and images, tool calls, and thinking, and scores 79.2% on SWE-Bench Verified).
ollama run inkling-2-mini
Note: inkling-2-mini is supported on Apple silicon only in this release. PC support will follow in an upcoming release.
What's Changed
-
ollama launch claudesupports Claude Code's deferred tool loading, so MCP tools are loaded when needed instead of all at once - Ollama on Apple silicon now serves multiple requests at once.
-
ollama launch opencodesets thinking options from the model's metadata -
ollama launch pirefreshes Pi's model list after sign-in and passes along prompt arguments -
ollama launch chatgpt --restoreasks before switching an active profile and preserves existing settings and chat history; quitting ChatGPT yourself now lets it exit through its own quit prompt -
Running
ollamawith no arguments in a non-interactive shell prints help instead of erroring -
Ctrl+C or Ctrl+G while in a multiline
"""block leaves multiline mode instead of swallowing the next line - Changing the context shift setting no longer reloads the model
-
Vulkan now uses mmap for integrated GPUs, fixing
Failed to allocate pinned memoryload failures on shared-memory GPUs - On Windows iGPUs, partially offloaded models no longer fail to allocate the KV cache
- Tool definitions passed through the API keep JSON schema keywords Ollama doesn't model instead of dropping them
- Disabling thinking now maps to the model's lowest supported reasoning level rather than being ignored
- Source builds can use prebuilt MLX-C and XGrammar binaries Full Changelog: v0.40.16...v0.40.17-rc3