Ollama releases: v0.34.1

TL;DR

Ollama released version 0.34.1, which marks the full integration of MLX safetensors support in their `ollama create` command and introduces stricter conditions for detecting runaway repeat tokens. These updates enhance model handling on Apple Silicon and improve overall stability.

Detailed Summary

Ollama released version 0.34.1, updating its MLX safetensors `ollama create` command to no longer be experimental and requiring the use of llama.cpp tooling for GGUF model creation. The update also enhances memory handling on Apple Silicon and introduces stricter conditions for detecting runaway repeat tokens. This release impacts users of Ollama's tools, particularly those working with model creation and management.

Key Points

• • MLX safetensors `ollama create` is no longer experimental.
• • GGUF model creation requires using llama.cpp tooling.
• • Improved MLX memory handling on Apple Silicon.
• • Runaway repeat token detection threshold increased to 100 repeats.