False
        

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

Release notes from ollama

2026-06-12T00:21:46Z tag:github.com,2008:Repository/658928958/v0.30.8 2026-06-12T19:57:22Z
2026-06-15T03:05:08Z tag:github.com,2008:Repository/658928958/v0.30.9-rc0 2026-06-15T03:05:08Z

v0.30.9-rc0

<p>llama: update llama.cpp to b9637 (<a class="issue-link js-issue-link" data-error-text="Failed to load title" data-id="4608333103" data-permission-text="Title is private" data-url="https://github.com/ollama/ollama/issues/16609" data-hovercard-type="pull_request" data-hovercard-url="/ollama/ollama/pull/16609/hovercard" href="https://github.com/ollama/ollama/pull/16609">#16609</a>)</p> jmorganca tag:github.com,2008:Repository/658928958/v0.30.8 2026-06-12T19:57:22Z

v0.30.8

<h2>What's Changed</h2> <ul> <li>Fixed <code>ollama launch</code> selecting the wrong provider in some cases</li> <li>Improved prompt caching by decoupling it from context shift for better KV cache reuse</li> <li>More stable MLX inference with hardened linear and embedding layers</li> <li>MLX runner now creates snapshots during prompt processing and speculative decoding for improved reliability</li> <li>Improved recurrent model support with per-boundary states from the gated-delta kernels</li> </ul> <p><strong>Full Changelog</strong>: <a class="commit-link" href="https://github.com/ollama/ollama/compare/v0.30.7...v0.30.8"><tt>v0.30.7...v0.30.8</tt></a></p> github-actions[bot] tag:github.com,2008:Repository/658928958/v0.30.8-rc0 2026-06-12T00:21:46Z

v0.30.8-rc0

<p>launch: Fix launch provider drift (<a class="issue-link js-issue-link" data-error-text="Failed to load title" data-id="4644799841" data-permission-text="Title is private" data-url="https://github.com/ollama/ollama/issues/16683" data-hovercard-type="pull_request" data-hovercard-url="/ollama/ollama/pull/16683/hovercard" href="https://github.com/ollama/ollama/pull/16683">#16683</a>)</p> ParthSareen tag:github.com,2008:Repository/658928958/v0.30.7 2026-06-08T20:18:26Z

v0.30.7

<p>Ollama Launch now supports Hermes Desktop, a native desktop interface for the Hermes agent. Run it alongside your Hermes agent to get a visual interface for managing conversations, integrations, and messaging apps.</p> <div class="snippet-clipboard-content notranslate position-relative overflow-auto" data-snippet-clipboard-copy-content="ollama launch hermes-desktop"><pre class="notranslate"><code>ollama launch hermes-desktop </code></pre></div> <a target="_blank" rel="noopener noreferrer" href="https://private-user-images.githubusercontent.com/29360864/604676516-3b2292d8-9f94-4d32-9023-85772e6ab3f8.png?jwt=eyJ0eXAiOiJKV1QiLCJhbGciOiJIUzI1NiJ9.eyJpc3MiOiJnaXRodWIuY29tIiwiYXVkIjoicmF3LmdpdGh1YnVzZXJjb250ZW50LmNvbSIsImtleSI6ImtleTUiLCJleHAiOjE3ODEzNDIzMzgsIm5iZiI6MTc4MTM0MjAzOCwicGF0aCI6Ii8yOTM2MDg2NC82MDQ2NzY1MTYtM2IyMjkyZDgtOWY5NC00ZDMyLTkwMjMtODU3NzJlNmFiM2Y4LnBuZz9YLUFtei1BbGdvcml0aG09QVdTNC1ITUFDLVNIQTI1NiZYLUFtei1DcmVkZW50aWFsPUFLSUFWQ09EWUxTQTUzUFFLNFpBJTJGMjAyNjA2MTMlMkZ1cy1lYXN0LTElMkZzMyUyRmF3czRfcmVxdWVzdCZYLUFtei1EYXRlPTIwMjYwNjEzVDA5MTM1OFomWC1BbXotRXhwaXJlcz0zMDAmWC1BbXotU2lnbmF0dXJlPTkyZWRjYzU2MzEwNTI2N2JjNjhhYWQyZWNjZjU0ODc3NGYyYTdmM2Y4YTM4ZDRjNmQ0YzM3NTU0MThkNWVhMWImWC1BbXotU2lnbmVkSGVhZGVycz1ob3N0JnJlc3BvbnNlLWNvbnRlbnQtdHlwZT1pbWFnZSUyRnBuZyJ9.-k0-p_L-kV9c3wrY6ksmF31UQw1H5BIAIswhEFk1xZs"><img width="2000" height="1343" alt="image" src="https://private-user-images.githubusercontent.com/29360864/604676516-3b2292d8-9f94-4d32-9023-85772e6ab3f8.png?jwt=eyJ0eXAiOiJKV1QiLCJhbGciOiJIUzI1NiJ9.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.-k0-p_L-kV9c3wrY6ksmF31UQw1H5BIAIswhEFk1xZs" content-type-secured-asset="image/png" style="max-width: 100%; height: auto; max-height: 1343px;"></a> <p>What's Changed</p> <ul> <li>Hermes Desktop is now available via <code>ollama launch hermes-desktop</code> with native Windows configuration path support</li> <li>OpenAI-compatible API models list now aligns with available model tags</li> <li>Added documentation describing the llama.cpp update process</li> <li>Updated Zod schema examples to use the native toJSONSchema helper</li> </ul> <p>Full Changelog: <a class="commit-link" href="https://github.com/ollama/ollama/compare/v0.30.6...v0.30.7"><tt>v0.30.6...v0.30.7</tt></a></p> github-actions[bot] tag:github.com,2008:Repository/658928958/v0.30.7-rc1 2026-06-06T00:59:05Z
<p>Ollama Launch now supports Hermes Desktop, a native desktop interface for the Hermes agent. Run it alongside your Hermes agent to get a visual interface for managing conversations, integrations, and messaging apps.</p> <div class="snippet-clipboard-content notranslate position-relative overflow-auto" data-snippet-clipboard-copy-content="ollama launch hermes-desktop"><pre class="notranslate"><code>ollama launch hermes-desktop </code></pre></div> <a target="_blank" rel="noopener noreferrer" href="https://private-user-images.githubusercontent.com/29360864/604676516-3b2292d8-9f94-4d32-9023-85772e6ab3f8.png?jwt=eyJ0eXAiOiJKV1QiLCJhbGciOiJIUzI1NiJ9.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.Z_a1fAQ55vGn-B-tbVsAernOJZSM6dBDQIStjShPmhg"><img width="2000" height="1343" alt="image" src="https://private-user-images.githubusercontent.com/29360864/604676516-3b2292d8-9f94-4d32-9023-85772e6ab3f8.png?jwt=eyJ0eXAiOiJKV1QiLCJhbGciOiJIUzI1NiJ9.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.Z_a1fAQ55vGn-B-tbVsAernOJZSM6dBDQIStjShPmhg" content-type-secured-asset="image/png" style="max-width: 100%; height: auto; max-height: 1343px;"></a> <p>What's Changed</p> <ul> <li>Hermes Desktop is now available via <code>ollama launch hermes-desktop</code> with native Windows configuration path support</li> <li>OpenAI-compatible API models list now aligns with available model tags</li> <li>Added documentation describing the llama.cpp update process</li> <li>Updated Zod schema examples to use the native toJSONSchema helper</li> </ul> <p>Full Changelog: <a class="commit-link" href="https://github.com/ollama/ollama/compare/v0.30.6...v0.30.7"><tt>v0.30.6...v0.30.7</tt></a></p> github-actions[bot] tag:github.com,2008:Repository/658928958/v0.30.7-rc1 2026-06-06T00:59:05Z

v0.30.7-rc1

<p>openai: align models list with tags (<a class="issue-link js-issue-link" data-error-text="Failed to load title" data-id="4600192735" data-permission-text="Title is private" data-url="https://github.com/ollama/ollama/issues/16556" data-hovercard-type="pull_request" data-hovercard-url="/ollama/ollama/pull/16556/hovercard" href="https://github.com/ollama/ollama/pull/16556">#16556</a>)</p> ParthSareen tag:github.com,2008:Repository/658928958/v0.30.7-rc0 2026-06-06T00:29:19Z

v0.30.7-rc0

<p>launch: use native Windows Hermes config path (<a class="issue-link js-issue-link" data-error-text="Failed to load title" data-id="4600810652" data-permission-text="Title is private" data-url="https://github.com/ollama/ollama/issues/16558" data-hovercard-type="pull_request" data-hovercard-url="/ollama/ollama/pull/16558/hovercard" href="https://github.com/ollama/ollama/pull/16558">#16558</a>)</p> BruceMacD tag:github.com,2008:Repository/658928958/v0.30.6 2026-06-07T00:12:11Z

v0.30.6

<h1>New models</h1> <ul> <li><a href="https://ollama.com/library/gemma4" rel="nofollow">Gemma 4 QAT weights</a>: the Gemma 4 family is now optimized with Quantization-Aware Training (QAT) to dramatically reduce memory requirements and maximize on-device performance. Look for the tags ending in <code>-qat</code>: <ul> <li><code>gemma4:e2b-it-qat</code></li> <li><code>gemma4:e4b-it-qat</code></li> <li><code>gemma4:12b-it-qat</code></li> <li><code>gemma4:26b-a4b-it-qat</code></li> <li><code>gemma4:31b-it-qat</code></li> </ul> </li> </ul> <h2>What's Changed</h2> <ul> <li><code>ollama launch omp</code> now integrates with <a href="https://omp.sh" rel="nofollow">Oh My Pi</a>, an AI coding agent with IDE integration</li> <li>MLX embedding layers now use NVFP4 global scale for improved quantization on Apple Silicon</li> </ul> <p><strong>Full Changelog</strong>: <a class="commit-link" href="https://github.com/ollama/ollama/compare/v0.30.5...v0.30.6"><tt>v0.30.5...v0.30.6</tt></a></p> github-actions[bot] tag:github.com,2008:Repository/658928958/v0.30.6-rc0 2026-06-05T00:49:49Z

v0.30.6-rc0

<p>launch: oh-my-pi (<a class="issue-link js-issue-link" data-error-text="Failed to load title" data-id="4567311798" data-permission-text="Title is private" data-url="https://github.com/ollama/ollama/issues/16410" data-hovercard-type="pull_request" data-hovercard-url="/ollama/ollama/pull/16410/hovercard" href="https://github.com/ollama/ollama/pull/16410">#16410</a>)</p> BruceMacD tag:github.com,2008:Repository/658928958/v0.30.5 2026-06-07T00:09:50Z

v0.30.5

<h2>What's Changed</h2> <ul> <li>Fixed the <code>gemma4:12b</code> floating point exception crash on x86, CUDA, Linux, and Windows systems.</li> <li><code>ollama launch hermes-desktop</code> now launches Hermes Desktop and can skip rebuilding when a packaged desktop app is already installed.</li> <li><code>ollama launch hermes</code> now supports native Windows installs through the Hermes PowerShell installer.</li> <li>Added Cline CLI integration docs.</li> </ul> <p><strong>Full Changelog</strong>: <a class="commit-link" href="https://github.com/ollama/ollama/compare/v0.30.4...v0.30.5"><tt>v0.30.4...v0.30.5</tt></a></p> github-actions[bot] tag:github.com,2008:Repository/658928958/v0.30.5-rc0 2026-06-04T15:20:57Z

v0.30.5-rc0: llama.cpp version update (#16511)

<p>Bump llama.cpp to b9509, which includes the upstream Gemma 4 12B multimodal projector fixes for the n_head=0 divide-by-zero crash seen on x86/CUDA/Linux/Windows.</p> <p>Fixes <a class="issue-link js-issue-link" data-error-text="Failed to load title" data-id="4583882035" data-permission-text="Title is private" data-url="https://github.com/ollama/ollama/issues/16479" data-hovercard-type="issue" data-hovercard-url="/ollama/ollama/issues/16479/hovercard" href="https://github.com/ollama/ollama/issues/16479">#16479</a><br> Fixes <a class="issue-link js-issue-link" data-error-text="Failed to load title" data-id="4584826238" data-permission-text="Title is private" data-url="https://github.com/ollama/ollama/issues/16489" data-hovercard-type="issue" data-hovercard-url="/ollama/ollama/issues/16489/hovercard" href="https://github.com/ollama/ollama/issues/16489">#16489</a><br> Fixes <a class="issue-link js-issue-link" data-error-text="Failed to load title" data-id="4585141411" data-permission-text="Title is private" data-url="https://github.com/ollama/ollama/issues/16491" data-hovercard-type="issue" data-hovercard-url="/ollama/ollama/issues/16491/hovercard" href="https://github.com/ollama/ollama/issues/16491">#16491</a><br> Fixes <a class="issue-link js-issue-link" data-error-text="Failed to load title" data-id="4585429816" data-permission-text="Title is private" data-url="https://github.com/ollama/ollama/issues/16492" data-hovercard-type="issue" data-hovercard-url="/ollama/ollama/issues/16492/hovercard" href="https://github.com/ollama/ollama/issues/16492">#16492</a><br> Fixes <a class="issue-link js-issue-link" data-error-text="Failed to load title" data-id="4585962519" data-permission-text="Title is private" data-url="https://github.com/ollama/ollama/issues/16495" data-hovercard-type="issue" data-hovercard-url="/ollama/ollama/issues/16495/hovercard" href="https://github.com/ollama/ollama/issues/16495">#16495</a></p> dhiltgen tag:github.com,2008:Repository/658928958/v0.30.4 2026-06-07T00:10:26Z

v0.30.4

<h1>New models</h1> <ul> <li><a href="https://ollama.com/library/nemotron-3-ultra" rel="nofollow">Nemotron-3-Ultra</a>: NVIDIA Nemotron 3 Ultra is built for high-throughput reasoning and long-running agent workflows.</li> </ul> <h2>What's Changed</h2> <ul> <li>Fixed multimodal models not using GPU on the llama.cpp backend can now use Metal GPU offload on Apple Silicon, improving multimodal performance on supported Macs.</li> <li><code>ollama create --experimental</code> now respects <code>REQUIRES</code> in Modelfiles for MLX-based models.</li> <li><code>ollama launch codex</code> now cleans up old conflicting Codex profile config before launching.</li> <li><code>ollama launch pi</code> now migrates users from the legacy Pi package to the official package and preserves the correct npm install prefix.</li> <li>Pi web search setup now updates only when a newer package is available.</li> <li>Windows cleanup now terminates the llama.cpp backend more reliably.</li> <li>Updated the llama.cpp backend.</li> </ul> <h2>Known Issues</h2> <ul> <li><code>gemma4:12b</code> crashes 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]
<p>Bump llama.cpp to b9509, which includes the upstream Gemma 4 12B multimodal projector fixes for the n_head=0 divide-by-zero crash seen on x86/CUDA/Linux/Windows.</p> <p>Fixes <a class="issue-link js-issue-link" data-error-text="Failed to load title" data-id="4583882035" data-permission-text="Title is private" data-url="https://github.com/ollama/ollama/issues/16479" data-hovercard-type="issue" data-hovercard-url="/ollama/ollama/issues/16479/hovercard" href="https://github.com/ollama/ollama/issues/16479">#16479</a><br> Fixes <a class="issue-link js-issue-link" data-error-text="Failed to load title" data-id="4584826238" data-permission-text="Title is private" data-url="https://github.com/ollama/ollama/issues/16489" data-hovercard-type="issue" data-hovercard-url="/ollama/ollama/issues/16489/hovercard" href="https://github.com/ollama/ollama/issues/16489">#16489</a><br> Fixes <a class="issue-link js-issue-link" data-error-text="Failed to load title" data-id="4585141411" data-permission-text="Title is private" data-url="https://github.com/ollama/ollama/issues/16491" data-hovercard-type="issue" data-hovercard-url="/ollama/ollama/issues/16491/hovercard" href="https://github.com/ollama/ollama/issues/16491">#16491</a><br> Fixes <a class="issue-link js-issue-link" data-error-text="Failed to load title" data-id="4585429816" data-permission-text="Title is private" data-url="https://github.com/ollama/ollama/issues/16492" data-hovercard-type="issue" data-hovercard-url="/ollama/ollama/issues/16492/hovercard" href="https://github.com/ollama/ollama/issues/16492">#16492</a><br> Fixes <a class="issue-link js-issue-link" data-error-text="Failed to load title" data-id="4585962519" data-permission-text="Title is private" data-url="https://github.com/ollama/ollama/issues/16495" data-hovercard-type="issue" data-hovercard-url="/ollama/ollama/issues/16495/hovercard" href="https://github.com/ollama/ollama/issues/16495">#16495</a></p> dhiltgen