Ollama 0.40.8-rc0
Ollama · App · v0.40.8-rc0: mlx: add Flint model support (#19181) · Released
View official release notes ↗ Copy link
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…
Source: Ollama's official Ollama release notes at github.com. Text reproduced from the source for reference; all rights belong to Ollama. Patchwire is not affiliated with Ollama.
← Older: v0.40.7Newer: v0.40.9-rc0: ci: switch to released WoA CTK… →
Get new Ollama releases by email
Every AI dev-tool release, including Ollama, in one short email each morning. Free. The digest is launching soon; join the list to get it from day one.
By subscribing you agree to our Terms and Privacy Policy. Unsubscribe anytime.
Nearby Ollama releases
- v0.40.9-rc2: flint: fix replies with thinking disabled (#19216) · Oct 22, 2026
- v0.40.9 · Oct 22, 2026
- v0.40.8 · Oct 21, 2026
- v0.40.9-rc0: ci: switch to released WoA CTK (#19185) · Oct 21, 2026
- v0.40.7 · Oct 21, 2026
- v0.40.7-rc1: mlx: deduplicate CUDA runtime payloads (#17956) · Oct 20, 2026
- v0.40.7 · Oct 18, 2026
- v0.40.7-rc0: mlx: add PPLX Decider v1.2 support (#19072) · Oct 17, 2026