v0.40.8-rc0: mlx: add Flint model support (#19181)
- Ollama: 87 events in the last 90 days
- Ollama: 81th Release in the last 90 days
- Previous: earlier the same day · v0.40.7-rc1: mlx: deduplicate CUDA runtime payloads (#17956)
What happened
Preserve source tool JSON alongside decoded API values so Flint can retain numeric types and schema order in prompts. Existing API marshaling is unchanged. Assemble ModelOpt expert banks during safetensors import, and reject direct Flint NVFP4 creation before conversion. Other models retain their existing import and quantization paths. Add optional learned sinks and device causal masks to shared MLX attention. Exist…
Summary assembled by rule from the sources below
Why it's spreading
Sources
Release
- v0.40.8-rc0: mlx: add Flint model support (#19181)Ollama Releases