tag:github.com,2008:https://github.com/ollama/ollama/releases

Release notes from ollama

2026-06-03T21:41:03Z tag:github.com,2008:Repository/658928958/v0.30.4 2026-06-04T01:08:55Z

v0.30.4

<h2>What's Changed</h2> <ul> <li>llama.cpp version update by <a class="user-mention notranslate" data-hovercard-type="user" data-hovercard-url="/users/dhiltgen/hovercard" data-octo-click="hovercard-link-click" data-octo-dimensions="link_type:self" href="https://github.com/dhiltgen">@dhiltgen</a> in <a class="issue-link js-issue-link" data-error-text="Failed to load title" data-id="4581515915" data-permission-text="Title is private" data-url="https://github.com/ollama/ollama/issues/16463" data-hovercard-type="pull_request" data-hovercard-url="/ollama/ollama/pull/16463/hovercard" href="https://github.com/ollama/ollama/pull/16463">#16463</a></li> <li>Kill llama-server during Windows cleanup by <a class="user-mention notranslate" data-hovercard-type="user" data-hovercard-url="/users/dhiltgen/hovercard" data-octo-click="hovercard-link-click" data-octo-dimensions="link_type:self" href="https://github.com/dhiltgen">@dhiltgen</a> in <a class="issue-link js-issue-link" data-error-text="Failed to load title" data-id="4580861003" data-permission-text="Title is private" data-url="https://github.com/ollama/ollama/issues/16458" data-hovercard-type="pull_request" data-hovercard-url="/ollama/ollama/pull/16458/hovercard" href="https://github.com/ollama/ollama/pull/16458">#16458</a></li> </ul> <h2>Known Issues</h2> <ul> <li>gemma4:12b crash with floating point exception</li> </ul> <p><strong>Full Changelog</strong>: <a class="commit-link" href="https://github.com/ollama/ollama/compare/v0.30.3...v0.30.4"><tt>v0.30.3...v0.30.4</tt></a></p> github-actions[bot] tag:github.com,2008:Repository/658928958/v0.30.4-rc1 2026-06-03T21:41:03Z

v0.30.4-rc1: llama-server: fix gemma4 patch wiring (#16477)

<p>This will fix the "clip.cpp:4399: Unknown projector type" crash.</p> dhiltgen tag:github.com,2008:Repository/658928958/v0.30.4-rc0 2026-06-03T17:25:12Z

v0.30.4-rc0: Kill llama-server during Windows cleanup (#16458)

<p>Windows installer and app cleanup could leave llama-server.exe running when ollama.exe was killed directly, so cleanup now includes llama-server.exe and taskkill /T.</p> dhiltgen tag:github.com,2008:Repository/658928958/v0.30.3 2026-06-03T16:39:44Z

v0.30.3

<h2>What's Changed</h2> <ul> <li>models: add support for gemma4-12b by <a class="user-mention notranslate" data-hovercard-type="user" data-hovercard-url="/users/pdevine/hovercard" data-octo-click="hovercard-link-click" data-octo-dimensions="link_type:self" href="https://github.com/pdevine">@pdevine</a> in <a class="issue-link js-issue-link" data-error-text="Failed to load title" data-id="4580836138" data-permission-text="Title is private" data-url="https://github.com/ollama/ollama/issues/16457" data-hovercard-type="pull_request" data-hovercard-url="/ollama/ollama/pull/16457/hovercard" href="https://github.com/ollama/ollama/pull/16457">#16457</a></li> </ul> <p><strong>Full Changelog</strong>: <a class="commit-link" href="https://github.com/ollama/ollama/compare/v0.30.2...v0.30.3"><tt>v0.30.2...v0.30.3</tt></a></p> github-actions[bot] tag:github.com,2008:Repository/658928958/v0.30.2 2026-06-03T02:54:30Z

v0.30.2

<h2>What's Changed</h2> <ul> <li>feat(launch): show and auto-install Cline CLI by <a class="user-mention notranslate" data-hovercard-type="user" data-hovercard-url="/users/hoyyeva/hovercard" data-octo-click="hovercard-link-click" data-octo-dimensions="link_type:self" href="https://github.com/hoyyeva">@hoyyeva</a> in <a class="issue-link js-issue-link" data-error-text="Failed to load title" data-id="4566649238" data-permission-text="Title is private" data-url="https://github.com/ollama/ollama/issues/16402" data-hovercard-type="pull_request" data-hovercard-url="/ollama/ollama/pull/16402/hovercard" href="https://github.com/ollama/ollama/pull/16402">#16402</a></li> <li>log template details to aid troubleshooting by <a class="user-mention notranslate" data-hovercard-type="user" data-hovercard-url="/users/dhiltgen/hovercard" data-octo-click="hovercard-link-click" data-octo-dimensions="link_type:self" href="https://github.com/dhiltgen">@dhiltgen</a> in <a class="issue-link js-issue-link" data-error-text="Failed to load title" data-id="4566872800" data-permission-text="Title is private" data-url="https://github.com/ollama/ollama/issues/16403" data-hovercard-type="pull_request" data-hovercard-url="/ollama/ollama/pull/16403/hovercard" href="https://github.com/ollama/ollama/pull/16403">#16403</a></li> <li>cmd/launch: add Qwen code integration by <a class="user-mention notranslate" data-hovercard-type="user" data-hovercard-url="/users/hoyyeva/hovercard" data-octo-click="hovercard-link-click" data-octo-dimensions="link_type:self" href="https://github.com/hoyyeva">@hoyyeva</a> in <a class="issue-link js-issue-link" data-error-text="Failed to load title" data-id="4359277115" data-permission-text="Title is private" data-url="https://github.com/ollama/ollama/issues/15900" data-hovercard-type="pull_request" data-hovercard-url="/ollama/ollama/pull/15900/hovercard" href="https://github.com/ollama/ollama/pull/15900">#15900</a></li> <li>launch: fix opencode local model limits by <a class="user-mention notranslate" data-hovercard-type="user" data-hovercard-url="/users/dhiltgen/hovercard" data-octo-click="hovercard-link-click" data-octo-dimensions="link_type:self" href="https://github.com/dhiltgen">@dhiltgen</a> in <a class="issue-link js-issue-link" data-error-text="Failed to load title" data-id="4572555638" data-permission-text="Title is private" data-url="https://github.com/ollama/ollama/issues/16425" data-hovercard-type="pull_request" data-hovercard-url="/ollama/ollama/pull/16425/hovercard" href="https://github.com/ollama/ollama/pull/16425">#16425</a></li> <li>llm: include cached prompt tokens in llama-server counts by <a class="user-mention notranslate" data-hovercard-type="user" data-hovercard-url="/users/dhiltgen/hovercard" data-octo-click="hovercard-link-click" data-octo-dimensions="link_type:self" href="https://github.com/dhiltgen">@dhiltgen</a> in <a class="issue-link js-issue-link" data-error-text="Failed to load title" data-id="4573277796" data-permission-text="Title is private" data-url="https://github.com/ollama/ollama/issues/16428" data-hovercard-type="pull_request" data-hovercard-url="/ollama/ollama/pull/16428/hovercard" href="https://github.com/ollama/ollama/pull/16428">#16428</a></li> <li>Harden app markdown URL handling by <a class="user-mention notranslate" data-hovercard-type="user" data-hovercard-url="/users/dhiltgen/hovercard" data-octo-click="hovercard-link-click" data-octo-dimensions="link_type:self" href="https://github.com/dhiltgen">@dhiltgen</a> in <a class="issue-link js-issue-link" data-error-text="Failed to load title" data-id="4559071258" data-permission-text="Title is private" data-url="https://github.com/ollama/ollama/issues/16380" data-hovercard-type="pull_request" data-hovercard-url="/ollama/ollama/pull/16380/hovercard" href="https://github.com/ollama/ollama/pull/16380">#16380</a></li> <li>discover: allow Radeon 8060S iGPU by default by <a class="user-mention notranslate" data-hovercard-type="user" data-hovercard-url="/users/dhiltgen/hovercard" data-octo-click="hovercard-link-click" data-octo-dimensions="link_type:self" href="https://github.com/dhiltgen">@dhiltgen</a> in <a class="issue-link js-issue-link" data-error-text="Failed to load title" data-id="4573365844" data-permission-text="Title is private" data-url="https://github.com/ollama/ollama/issues/16429" data-hovercard-type="pull_request" data-hovercard-url="/ollama/ollama/pull/16429/hovercard" href="https://github.com/ollama/ollama/pull/16429">#16429</a></li> <li>llm: detect llama-server load stalls from output by <a class="user-mention notranslate" data-hovercard-type="user" data-hovercard-url="/users/dhiltgen/hovercard" data-octo-click="hovercard-link-click" data-octo-dimensions="link_type:self" href="https://github.com/dhiltgen">@dhiltgen</a> in <a class="issue-link js-issue-link" data-error-text="Failed to load title" data-id="4573087391" data-permission-text="Title is private" data-url="https://github.com/ollama/ollama/issues/16427" data-hovercard-type="pull_request" data-hovercard-url="/ollama/ollama/pull/16427/hovercard" href="https://github.com/ollama/ollama/pull/16427">#16427</a></li> <li>More harden app markdown URL handling by <a class="user-mention notranslate" data-hovercard-type="user" data-hovercard-url="/users/dhiltgen/hovercard" data-octo-click="hovercard-link-click" data-octo-dimensions="link_type:self" href="https://github.com/dhiltgen">@dhiltgen</a> in <a class="issue-link js-issue-link" data-error-text="Failed to load title" data-id="4574048980" data-permission-text="Title is private" data-url="https://github.com/ollama/ollama/issues/16436" data-hovercard-type="pull_request" data-hovercard-url="/ollama/ollama/pull/16436/hovercard" href="https://github.com/ollama/ollama/pull/16436">#16436</a></li> <li>llama.cpp version update by <a class="user-mention notranslate" data-hovercard-type="user" data-hovercard-url="/users/dhiltgen/hovercard" data-octo-click="hovercard-link-click" data-octo-dimensions="link_type:self" href="https://github.com/dhiltgen">@dhiltgen</a> in <a class="issue-link js-issue-link" data-error-text="Failed to load title" data-id="4572890762" data-permission-text="Title is private" data-url="https://github.com/ollama/ollama/issues/16426" data-hovercard-type="pull_request" data-hovercard-url="/ollama/ollama/pull/16426/hovercard" href="https://github.com/ollama/ollama/pull/16426">#16426</a></li> <li>launch: isolate Codex launch configuration by <a class="user-mention notranslate" data-hovercard-type="user" data-hovercard-url="/users/ParthSareen/hovercard" data-octo-click="hovercard-link-click" data-octo-dimensions="link_type:self" href="https://github.com/ParthSareen">@ParthSareen</a> in <a class="issue-link js-issue-link" data-error-text="Failed to load title" data-id="4574163252" data-permission-text="Title is private" data-url="https://github.com/ollama/ollama/issues/16437" data-hovercard-type="pull_request" data-hovercard-url="/ollama/ollama/pull/16437/hovercard" href="https://github.com/ollama/ollama/pull/16437">#16437</a></li> <li>llama: add laguna (poolside) arch via a llama.cpp patch under llama/c… by <a class="user-mention notranslate" data-hovercard-type="user" data-hovercard-url="/users/dhiltgen/hovercard" data-octo-click="hovercard-link-click" data-octo-dimensions="link_type:self" href="https://github.com/dhiltgen">@dhiltgen</a> in <a class="issue-link js-issue-link" data-error-text="Failed to load title" data-id="4565408846" data-permission-text="Title is private" data-url="https://github.com/ollama/ollama/issues/16396" data-hovercard-type="pull_request" data-hovercard-url="/ollama/ollama/pull/16396/hovercard" href="https://github.com/ollama/ollama/pull/16396">#16396</a></li> <li>docs: configure hermes desktop app by <a class="user-mention notranslate" data-hovercard-type="user" data-hovercard-url="/users/BruceMacD/hovercard" data-octo-click="hovercard-link-click" data-octo-dimensions="link_type:self" href="https://github.com/BruceMacD">@BruceMacD</a> in <a class="issue-link js-issue-link" data-error-text="Failed to load title" data-id="4575033150" data-permission-text="Title is private" data-url="https://github.com/ollama/ollama/issues/16440" data-hovercard-type="pull_request" data-hovercard-url="/ollama/ollama/pull/16440/hovercard" href="https://github.com/ollama/ollama/pull/16440">#16440</a></li> <li>llm: ignore llama-server SSE ping comments by <a class="user-mention notranslate" data-hovercard-type="user" data-hovercard-url="/users/dhiltgen/hovercard" data-octo-click="hovercard-link-click" data-octo-dimensions="link_type:self" href="https://github.com/dhiltgen">@dhiltgen</a> in <a class="issue-link js-issue-link" data-error-text="Failed to load title" data-id="4575429520" data-permission-text="Title is private" data-url="https://github.com/ollama/ollama/issues/16443" data-hovercard-type="pull_request" data-hovercard-url="/ollama/ollama/pull/16443/hovercard" href="https://github.com/ollama/ollama/pull/16443">#16443</a></li> <li>fix laguna patch build breakage by <a class="user-mention notranslate" data-hovercard-type="user" data-hovercard-url="/users/dhiltgen/hovercard" data-octo-click="hovercard-link-click" data-octo-dimensions="link_type:self" href="https://github.com/dhiltgen">@dhiltgen</a> in <a class="issue-link js-issue-link" data-error-text="Failed to load title" data-id="4575632103" data-permission-text="Title is private" data-url="https://github.com/ollama/ollama/issues/16445" data-hovercard-type="pull_request" data-hovercard-url="/ollama/ollama/pull/16445/hovercard" href="https://github.com/ollama/ollama/pull/16445">#16445</a></li> </ul> <p><strong>Full Changelog</strong>: <a class="commit-link" href="https://github.com/ollama/ollama/compare/v0.30.0...v0.30.2-rc0"><tt>v0.30.0...v0.30.2-rc0</tt></a></p> github-actions[bot] tag:github.com,2008:Repository/658928958/v0.30.2-rc0 2026-06-02T23:35:19Z

v0.30.2-rc0: fix laguna patch build breakage (#16445)

<p>Follow up to <a class="issue-link js-issue-link" data-error-text="Failed to load title" data-id="4565408846" data-permission-text="Title is private" data-url="https://github.com/ollama/ollama/issues/16396" data-hovercard-type="pull_request" data-hovercard-url="/ollama/ollama/pull/16396/hovercard" href="https://github.com/ollama/ollama/pull/16396">#16396</a></p> <p>Fix kernel template instantiation so the symbols are exported in the library.</p> dhiltgen tag:github.com,2008:Repository/658928958/v0.30.1 2026-06-02T22:40:14Z

v0.30.1: llm: ignore llama-server SSE ping comments (#16443)

<p>llama.cpp b9478 added a default 30s SSE ping that emits colon-only comment frames (":\n\n") while streamed requests are idle; Ollama treated non-data SSE lines as JSON, so skip SSE comments in completion and chat streams.</p> dhiltgen tag:github.com,2008:Repository/658928958/v0.30.1-rc0 2026-06-02T19:10:46Z

v0.30.1-rc0

<p>launch: isolate Codex launch configuration (<a class="issue-link js-issue-link" data-error-text="Failed to load title" data-id="4574163252" data-permission-text="Title is private" data-url="https://github.com/ollama/ollama/issues/16437" data-hovercard-type="pull_request" data-hovercard-url="/ollama/ollama/pull/16437/hovercard" href="https://github.com/ollama/ollama/pull/16437">#16437</a>)</p> ParthSareen tag:github.com,2008:Repository/658928958/v0.30.0 2026-06-02T00:32:00Z

v0.30.0

<p>Ollama 0.30 is now available, with improved compatibility and performance using <a href="https://github.com/ggml-org/llama.cpp">llama.cpp</a>. This augments the MLX engine on Apple Silicon, bringing support to a wider range of hardware.</p> <p>This release brings support for a wider range of models, including GGUF-based models from Hugging Face and your own fine-tuned models along with faster performance on NVIDIA hardware.</p> <h2>Known issues:</h2> <ul> <li><code>laguna-xs.2</code> is not yet supported on Windows/Linux.</li> <li><code>llama3.2-vision</code> is not yet supported</li> <li><code>nomic-embed-text</code> now converts inputs to lowercase per the model card where prior Ollama versions incorrectly preserved mixed case</li> </ul> github-actions[bot] tag:github.com,2008:Repository/658928958/v0.30.0-rc32 2026-06-01T17:44:21Z

v0.30.0-rc32: llama-server followups (#16353)

<ul> <li>llama-server followups</li> </ul> <p>Misc fixes for <a class="issue-link js-issue-link" data-error-text="Failed to load title" data-id="4395240930" data-permission-text="Title is private" data-url="https://github.com/ollama/ollama/issues/16031" data-hovercard-type="pull_request" data-hovercard-url="/ollama/ollama/pull/16031/hovercard" href="https://github.com/ollama/ollama/pull/16031">#16031</a></p> <ul> <li>Add back dropped ROCm build flag for multi-GPU support on windows</li> <li>Fix amdhip64_*.dll version detection for "latest" selection</li> <li>Fix embeddings API for consistent normalize behavior with prior versions</li> </ul> <ul> <li> <p>ci: set up for automated llama.cpp update testing</p> </li> <li> <p>reduce batch for fa-disabled, and constrained vram</p> </li> <li> <p>mlx: fix v3 load bug on m5</p> </li> </ul> <p>Imagegen was incorrectly loading v3 first. This DRYs out the loading code so imagegen gets the same new v4/v3 selection logic.</p> <ul> <li> <p>fix reload bug on embedding models</p> </li> <li> <p>bump version</p> </li> <li> <p>steer user how to enable iGPU when disabled</p> </li> </ul> dhiltgen