Skip to content
Live
Trending storyPast story

Ollama added support for Flint model

1 reports1 outlets28 days agoUpdate

The story so far

AI overview

Ollama released version v0.40.8-rc0, adding support for the Flint model. By retaining the source tool JSON and decoding API values, Flint maintains numeric types and pattern order in prompts while keeping the existing API serialization unchanged. Additionally, the ModelOpt expert library is assembled during safetensors import, and direct creation of Flint NVFP4 is rejected before conversion. Other models retain their existing import and quantization paths. Optional learning receivers and device causal masks are added to shared MLX attention, with existing callers retaining no receiver and host mask paths; Flint adopts the new behavior.

AI-written from the reports · Updated 27 days ago

Coverage timeline

Follow the coverage to see every side of the story.

21 Oct
  1. Ollama releases
    v0.40.8-rc0: mlx: add Flint model support (#19181)

    Ollama v0.40.8-rc0 has been released, adding support for the Flint model. It retains the source tool JSON and decoding API values to maintain numeric types and pattern order in prompts, while keeping the existing API serialization unchanged.