--- title: IBM Granite 3.0 models · Ollama Blog | daily.dev description: IBM Granite 3.0 models are now available to run through Ollama under an Apache 2.0 license. The lineup includes 2B and 8B dense text-only LLMs trained on over... canonical: https://daily.dev/posts/ibm-granite-3-0-models-ollama-blog-jvlftup5v twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: IBM Granite 3.0 models · Ollama Blog | daily.dev og:description: IBM Granite 3.0 models are now available to run through Ollama under an Apache 2.0 license. The lineup includes 2B and 8B dense text-only LLMs trained on over... og:url: https://daily.dev/posts/ibm-granite-3-0-models-ollama-blog-jvlftup5v og:image: https://api.daily.dev/og/posts/jvLfTup5v.png og:image:alt: IBM Granite 3.0 models · Ollama Blog og:image:width: 1200 og:image:height: 630 og:locale: en --- > ## Documentation Index > Fetch the complete documentation index at: https://daily.dev/llms.txt > Use this file to discover all available pages before exploring further. # IBM Granite 3.0 models · Ollama Blog **[Ollama](https://daily.dev/sources/ollama)** · 2 min read · 0 upvotes · 0 comments ## Summary IBM Granite 3.0 models are now available to run through Ollama under an Apache 2.0 license. The lineup includes 2B and 8B dense text-only LLMs trained on over 12 trillion tokens, with Granite 8B Instruct rivaling Llama 3.1 8B Instruct on OpenLLM Leaderboard benchmarks. Also included are 1B and 3B mixture-of-experts (MoE) models trained on over 10 trillion tokens, designed for low-latency and on-device inference. The models support tool use, RAG, code generation, translation, bug fixing, summarization, text classification, extraction, question-answering, and multilingual dialog. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Questions this post answers ### How do I run IBM Granite 3.0 models locally with Ollama? Use the ollama run command with the specific model tag. For the 2B dense model run 'ollama run granite3-dense', for the 8B dense model run 'ollama run granite3-dense:8b', for the 1B MoE model run 'ollama run granite3-moe', and for the 3B MoE model run 'ollama run granite3-moe:3b'. All are licensed under Apache 2.0. _daily.dev helps developers track new local-model releases like Granite 3.0 as they roll out._ ### What is the difference between IBM Granite 3.0 dense models and MoE models? Granite 3.0 dense models (2B and 8B) are text-only LLMs trained on over 12 trillion tokens, built for general performance rivaling Llama 3.1 8B Instruct on OpenLLM Leaderboard benchmarks. The 1B and 3B MoE models are trained on over 10 trillion tokens and optimized for low latency, making them suited for on-device or instantaneous-inference scenarios. _Developers choosing between dense and MoE models can follow comparisons like this on daily.dev._ ### How does IBM Granite 8B Instruct compare to Llama 3.1 8B Instruct? Granite 8B Instruct rivals Llama 3.1 8B Instruct in performance across both OpenLLM Leaderboard v1 and v2 benchmarks, according to IBM's initial testing. It is a text-only dense LLM trained on over 12 trillion tokens, licensed under Apache 2.0 and runnable through Ollama. _daily.dev keeps developers evaluating open LLMs like Granite and Llama up to date on benchmark results._ ## Similar posts on daily.dev - [Running Granite 4 Language Models with Ollama](https://daily.dev/posts/running-granite-4-language-models-with-ollama-okn28odo9) · Niklas Heidloff · 1 upvotes · 0 comments - [New IBM Granite 4 Models to Reduce AI Costs with Inference-Efficient Hybrid Mamba-2 Architecture](https://daily.dev/posts/new-ibm-granite-4-models-to-reduce-ai-costs-with-inference-efficient-hybrid-mamba-2-architecture-nwshaxq1n) · InfoQ · 0 upvotes · 0 comments --- Tags: [#llm](https://daily.dev/tags/llm), [#ollama](https://daily.dev/tags/ollama), [#mixture-of-experts](https://daily.dev/tags/mixture-of-experts) [View this post on daily.dev](https://daily.dev/posts/ibm-granite-3-0-models-ollama-blog-jvlftup5v) ```json {"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://daily.dev/#organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180},"sameAs":["https://twitter.com/dailydotdev","https://github.com/dailydotdev","https://www.linkedin.com/company/daily-dev-ltd"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","publisher":{"@id":"https://daily.dev/#organization"},"potentialAction":{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https://daily.dev/search?q={search_term_string}"},"query-input":"required name=search_term_string"}}]} {"@context":"https://schema.org","@type":"TechArticle","headline":"IBM Granite 3.0 models · Ollama Blog","url":"https://daily.dev/posts/ibm-granite-3-0-models-ollama-blog-jvlftup5v","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/ibm-granite-3-0-models-ollama-blog-jvlftup5v"},"datePublished":"2026-08-23T12:22:21.278Z","dateModified":"2026-08-23T12:51:24.056Z","description":"IBM Granite 3.0 models are now available to run through Ollama under an Apache 2.0 license. The lineup includes 2B and 8B dense text-only LLMs trained on over...","image":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/03fb1a10ef158f8986c9778334c2af42?_a=AQAEuop","thumbnailUrl":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/03fb1a10ef158f8986c9778334c2af42?_a=AQAEuop","isAccessibleForFree":true,"articleSection":"Ollama","inLanguage":"en","publisher":{"@type":"Organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180}},"author":{"@type":"Organization","name":"Ollama","logo":"https://media.daily.dev/image/upload/s--dRYYkBCj--/f_auto,q_auto/v1787487632/logos/ollama-blog?_a=BAMAMicg0","url":"https://daily.dev/sources/ollama"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/ibm-granite-3-0-models-ollama-blog-jvlftup5v","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":0},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"llm,ollama,mixture-of-experts","timeRequired":"PT2M"} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"Ollama","item":"https://daily.dev/sources/ollama"},{"@type":"ListItem","position":3,"name":"IBM Granite 3.0 models · Ollama Blog"}]} {"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/ibm-granite-3-0-models-ollama-blog-jvlftup5v#faq","mainEntity":[{"@type":"Question","name":"How do I run IBM Granite 3.0 models locally with Ollama?","acceptedAnswer":{"@type":"Answer","text":"Use the ollama run command with the specific model tag. For the 2B dense model run 'ollama run granite3-dense', for the 8B dense model run 'ollama run granite3-dense:8b', for the 1B MoE model run 'ollama run granite3-moe', and for the 3B MoE model run 'ollama run granite3-moe:3b'. All are licensed under Apache 2.0. daily.dev helps developers track new local-model releases like Granite 3.0 as they roll out."}},{"@type":"Question","name":"What is the difference between IBM Granite 3.0 dense models and MoE models?","acceptedAnswer":{"@type":"Answer","text":"Granite 3.0 dense models (2B and 8B) are text-only LLMs trained on over 12 trillion tokens, built for general performance rivaling Llama 3.1 8B Instruct on OpenLLM Leaderboard benchmarks. The 1B and 3B MoE models are trained on over 10 trillion tokens and optimized for low latency, making them suited for on-device or instantaneous-inference scenarios. Developers choosing between dense and MoE models can follow comparisons like this on daily.dev."}},{"@type":"Question","name":"How does IBM Granite 8B Instruct compare to Llama 3.1 8B Instruct?","acceptedAnswer":{"@type":"Answer","text":"Granite 8B Instruct rivals Llama 3.1 8B Instruct in performance across both OpenLLM Leaderboard v1 and v2 benchmarks, according to IBM's initial testing. It is a text-only dense LLM trained on over 12 trillion tokens, licensed under Apache 2.0 and runnable through Ollama. daily.dev keeps developers evaluating open LLMs like Granite and Llama up to date on benchmark results."}}]} ```