> ## Content Index
> Fetch the complete content index at: https://globalfeed.ai/llms.txt
> Use this file to discover other available public pages before exploring further.

# Ollama fixes growing memory use during long responses
- URL: https://globalfeed.ai/en/ollama-fixes-growing-memory-use-during-long-responses-2d2129e3/
- Published: 2026-09-20T01:40:54.000Z
- Updated: 2026-09-20T01:40:54.000Z
- Description: According to Ollama's release notes, excessive memory growth during long responses with MLX has been fixed. The stated fix concerns generation using speculative decoding; it is not an announcement of a general speed increase across all models.
- Author: GlobalFeed Editor
- Tags: Ollama, Artificial Intelligence, x-ollama, dil-en, video, elle, #gf-autopilot, #gf-event-2d2129e34a831321cab8e3c4

[Primary source](https://github.com/ollama/ollama/releases/tag/v0.34.2?ref=globalfeed.ai)

According to Ollama's release notes, excessive memory growth during long responses with MLX has been fixed. The stated fix concerns generation using speculative decoding; it is not an announcement of a general speed increase across all models. [github.com](https://github.com/ollama/ollama/releases/tag/v0.34.2?ref=globalfeed.ai)

The development record explains the cause: the generation loop was supposed to release the freed-buffer pool at each 256-token boundary. Because speculative generation can emit several tokens in one round, the counter could step over the exact boundary and fail to trigger the release check. [github.com](https://github.com/ollama/ollama/compare/v0.34.1...v0.34.2?ref=globalfeed.ai)

The revised check releases the pool whenever a round crosses a multiple of 256 tokens. The counter no longer has to stop exactly on the boundary; the release behavior already used for single-token rounds now also covers rounds that step past it. [github.com](https://github.com/ollama/ollama/compare/v0.34.1...v0.34.2?ref=globalfeed.ai)

The same release adds a first-run setup when the Ollama command is launched, offering the choice to sign in or continue locally. Setup completion is shared with the desktop application on macOS and Windows. Continuing locally is explicitly listed as an option in the release notes. [github.com](https://github.com/ollama/ollama/releases/tag/v0.34.2?ref=globalfeed.ai)

This report is based on published technical documentation. GlobalFeed has not independently benchmarked the implementation.