[KongchangAI]
Person

dhiltgen

A maintainer of the Ollama project, responsible for version releases and code merges.

Timeline (last 90 days)

Oct 15

The commit corresponding to Ollama v0.40.5 is d919272, tagged by maintainer dhiltgen on October 15 at 16:11, with the commit message 'llama.cpp: version bump b11669 (#19032)'.

Unverified50%
Oct 15

The core change in Ollama v0.40.5 is porting MLX's Metal custom kernels to the CUDA platform (PR #18657, author @dhiltgen).

Unverified50%
Oct 14

Ollama v0.40.5-rc0 was tagged by maintainer dhiltgen, corresponding to commit 292d7ce, with the commit message "MLX: version bump (#18993)".

Unverified50%
Oct 14

Ollama v0.40.4 contains three changes directly related to MLX: state compaction after speculative rollback and an MLX / MLX-C version upgrade, both submitted by dhiltgen.

Unverified50%
Oct 14

In Ollama v0.40.4, dhiltgen submitted server-side changes focused on performance optimization of inference routing, reusing already-validated manifests and already-loaded models.

Unverified50%
Oct 13

This update was tagged and committed by maintainer dhiltgen, and carries a signature verification mark on GitHub.

Unverified50%
Oct 12

Ollama 发布了 v0.40.4-rc0 预发布版本,对应提交为 374d97d,由开发者 dhiltgen 打标签。

Unverified50%
Oct 8

Core maintainer @dhiltgen implemented hiding of duplicate entries and a downgrade guard in the server list in PR #18874.

Unverified50%
Oct 8

This version removes the metal residency patch from MLX that has been merged upstream (#18854, contributor dhiltgen).

Unverified50%
Oct 8

A fix submitted by dhiltgen (#18852) causes manifest handling to avoid symlinks on Windows.

Unverified50%

3 more timeline events

All Facts (13)

Unverified

The commit corresponding to Ollama v0.40.5 is d919272, tagged by maintainer dhiltgen on October 15 at 16:11, with the commit message 'llama.cpp: version bump b11669 (#19032)'.

50%
Unverified

The core change in Ollama v0.40.5 is porting MLX's Metal custom kernels to the CUDA platform (PR #18657, author @dhiltgen).

50%
Unverified

Ollama v0.40.5-rc0 was tagged by maintainer dhiltgen, corresponding to commit 292d7ce, with the commit message "MLX: version bump (#18993)".

50%
Unverified

This update was tagged and committed by maintainer dhiltgen, and carries a signature verification mark on GitHub.

50%
Unverified

A fix submitted by dhiltgen (#18852) causes manifest handling to avoid symlinks on Windows.

50%
Unverified

This version removes the metal residency patch from MLX that has been merged upstream (#18854, contributor dhiltgen).

50%
Unverified

v0.40.0-rc1 was tagged and released by project maintainer dhiltgen, with the commit carrying a GitHub-verified signature.

50%
Unverified

The v0.35.1-rc2 version was released by maintainer dhiltgen, with commit hash b0c1ca4.

50%
Unverified

In Ollama v0.40.4, dhiltgen submitted server-side changes focused on performance optimization of inference routing, reusing already-validated manifests and already-loaded models.

50%
Unverified

Ollama v0.40.4 contains three changes directly related to MLX: state compaction after speculative rollback and an MLX / MLX-C version upgrade, both submitted by dhiltgen.

50%
Unverified

Ollama 发布了 v0.40.4-rc0 预发布版本,对应提交为 374d97d,由开发者 dhiltgen 打标签。

50%
Unverified

Core maintainer @dhiltgen implemented hiding of duplicate entries and a downgrade guard in the server list in PR #18874.

50%
Unverified

Ollama released a preview version v0.40.0-rc4 (Release Candidate 4), tagged by maintainer dhiltgen.

50%

Source Articles