--- title: Thinking · Ollama Blog | daily.dev description: Ollama added the ability to enable or disable a model's 'thinking' output, letting developers separate reasoning traces from final answers. Supported models... canonical: https://daily.dev/posts/thinking-ollama-blog-yoclgzyld twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: Thinking · Ollama Blog | daily.dev og:description: Ollama added the ability to enable or disable a model's 'thinking' output, letting developers separate reasoning traces from final answers. Supported models... og:url: https://daily.dev/posts/thinking-ollama-blog-yoclgzyld og:image: https://api.daily.dev/og/posts/YoclGzyLD.png og:image:alt: Thinking · 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. # Thinking · Ollama Blog **[Ollama](https://daily.dev/sources/ollama)** · 4 min read · 0 upvotes · 0 comments ## Summary Ollama added the ability to enable or disable a model's 'thinking' output, letting developers separate reasoning traces from final answers. Supported models include DeepSeek R1 and Qwen 3, with more being added under a dedicated thinking-models category. The feature is available via CLI flags (--think, --think=false), interactive session commands (/set think, /set nothink), a --hidethinking scripting flag, and a new think parameter in the generate and chat APIs, plus updated Python and JavaScript libraries. ## 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 enable or disable thinking output for DeepSeek R1 in Ollama's CLI? Thinking is enabled by default in the CLI. To disable it interactively, use /set nothink followed by your prompt; to enable it again, use /set think. For non-interactive runs, pass --think or --think=false as command flags. A --hidethinking flag also lets you run a thinking model while only showing the final answer. _Developers wiring up reasoning models into their own tools can track Ollama's CLI changes on daily.dev._ ### How do I access the model's thinking output using Ollama's chat API? Set the think parameter to true in a request to /api/chat or /api/generate. The response then includes a separate thinking field alongside the usual content field, letting you display or animate the reasoning process independently from the final answer, for example in a graphical interface or as an NPC thought bubble in a game. _daily.dev helps developers building on evolving LLM APIs like Ollama's chat endpoint stay current._ ### How do I get the thinking output when using Ollama's Python or JavaScript client libraries? Pass think=True (Python) or think: true (JavaScript) to the chat function call, then read response.message.thinking separately from response.message.content. For streaming responses in JavaScript, check chunk.message.thinking versus chunk.message.content on each chunk to separate the reasoning stream from the final output stream. _daily.dev keeps developers integrating LLM SDKs like Ollama's Python and JS libraries up to date._ ## Similar posts on daily.dev - [Qwen3.5 in .NET MEAI Hangs Forever And How OllamaSharp Fixes It in One Line](https://daily.dev/posts/qwen3-5-in-net-meai-hangs-forever-and-how-ollamasharp-fixes-it-in-one-line-ircj5xn5e) · We Are .NET · 0 upvotes · 0 comments - [Building an Intelligent Chatbot with Qwen3 Instruct and Thinking Models](https://daily.dev/posts/building-an-intelligent-chatbot-with-qwen3-instruct-and-thinking-models-rjickxrqy) · PyImageSearch · 0 upvotes · 0 comments - [Setting Up Your Own Large Language Model](https://daily.dev/posts/setting-up-your-own-large-language-model-752lyoupu) · Towards Data Science · 1 upvotes · 0 comments - [I stopped treating AI models like Pokémon—I don't need to collect them all](https://daily.dev/posts/i-stopped-treating-ai-models-like-pok-mon-i-don-t-need-to-collect-them-all-0p7nvluvs) · XDA Developers · 1 upvotes · 0 comments --- Tags: [#llm](https://daily.dev/tags/llm), [#ollama](https://daily.dev/tags/ollama) [View this post on daily.dev](https://daily.dev/posts/thinking-ollama-blog-yoclgzyld) ```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":"Thinking · Ollama Blog","url":"https://daily.dev/posts/thinking-ollama-blog-yoclgzyld","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/thinking-ollama-blog-yoclgzyld"},"datePublished":"2026-08-23T12:23:51.263Z","dateModified":"2026-08-23T12:57:22.783Z","description":"Ollama added the ability to enable or disable a model's 'thinking' output, letting developers separate reasoning traces from final answers. Supported models...","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/thinking-ollama-blog-yoclgzyld","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":0},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"llm,ollama","timeRequired":"PT4M"} {"@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":"Thinking · Ollama Blog"}]} {"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/thinking-ollama-blog-yoclgzyld#faq","mainEntity":[{"@type":"Question","name":"How do I enable or disable thinking output for DeepSeek R1 in Ollama's CLI?","acceptedAnswer":{"@type":"Answer","text":"Thinking is enabled by default in the CLI. To disable it interactively, use /set nothink followed by your prompt; to enable it again, use /set think. For non-interactive runs, pass --think or --think=false as command flags. A --hidethinking flag also lets you run a thinking model while only showing the final answer. Developers wiring up reasoning models into their own tools can track Ollama's CLI changes on daily.dev."}},{"@type":"Question","name":"How do I access the model's thinking output using Ollama's chat API?","acceptedAnswer":{"@type":"Answer","text":"Set the think parameter to true in a request to /api/chat or /api/generate. The response then includes a separate thinking field alongside the usual content field, letting you display or animate the reasoning process independently from the final answer, for example in a graphical interface or as an NPC thought bubble in a game. daily.dev helps developers building on evolving LLM APIs like Ollama's chat endpoint stay current."}},{"@type":"Question","name":"How do I get the thinking output when using Ollama's Python or JavaScript client libraries?","acceptedAnswer":{"@type":"Answer","text":"Pass think=True (Python) or think: true (JavaScript) to the chat function call, then read response.message.thinking separately from response.message.content. For streaming responses in JavaScript, check chunk.message.thinking versus chunk.message.content on each chunk to separate the reasoning stream from the final output stream. daily.dev keeps developers integrating LLM SDKs like Ollama's Python and JS libraries up to date."}}]} ```