[{"data":1,"prerenderedAt":323},["ShallowReactive",2],{"docs-\u002Fdocs\u002Fv2\u002Fagent\u002Fengines\u002Follama":3},{"id":4,"title":5,"body":6,"description":316,"extension":317,"meta":318,"navigation":70,"path":319,"seo":320,"stem":321,"__hash__":322},"docs\u002Fdocs\u002Fv2\u002Fagent\u002Fengines\u002Follama.md","Ollama",{"type":7,"value":8,"toc":310},"minimark",[9,13,24,29,115,125,128,206,210,217,220,237,241,297,301,306],[10,11,5],"h1",{"id":12},"ollama",[14,15,16,17,23],"p",{},"Local inference on your machine. No API costs, no data leaving your machine. Requires\n",[18,19,5],"a",{"href":20,"rel":21},"https:\u002F\u002Follama.com",[22],"nofollow"," installed — Axon manages the server lifecycle automatically.",[25,26,28],"h2",{"id":27},"configuration","Configuration",[30,31,36],"pre",{"className":32,"code":33,"language":34,"meta":35,"style":35},"language-ts shiki shiki-themes dark-plus","import { Ollama } from \"@arcforge\u002Fengines\"\n\nexport default defineAgent({\n    engine: Ollama({ model: \"qwen2.5:7b\" }),\n})\n","ts","",[37,38,39,65,72,88,109],"code",{"__ignoreMap":35},[40,41,44,48,52,55,58,61],"span",{"class":42,"line":43},"line",1,[40,45,47],{"class":46},"sYbnZ","import",[40,49,51],{"class":50},"sTNBD"," { ",[40,53,5],{"class":54},"s3F5K",[40,56,57],{"class":50}," } ",[40,59,60],{"class":46},"from",[40,62,64],{"class":63},"sKc5r"," \"@arcforge\u002Fengines\"\n",[40,66,68],{"class":42,"line":67},2,[40,69,71],{"emptyLinePlaceholder":70},true,"\n",[40,73,75,78,81,85],{"class":42,"line":74},3,[40,76,77],{"class":46},"export",[40,79,80],{"class":46}," default",[40,82,84],{"class":83},"sCudf"," defineAgent",[40,86,87],{"class":50},"({\n",[40,89,91,94,97,100,103,106],{"class":42,"line":90},4,[40,92,93],{"class":54},"    engine:",[40,95,96],{"class":83}," Ollama",[40,98,99],{"class":50},"({ ",[40,101,102],{"class":54},"model:",[40,104,105],{"class":63}," \"qwen2.5:7b\"",[40,107,108],{"class":50}," }),\n",[40,110,112],{"class":42,"line":111},5,[40,113,114],{"class":50},"})\n",[14,116,117,120,121,124],{},[37,118,119],{},"model"," is the model name as it appears in ",[37,122,123],{},"ollama list",". Any model you've pulled is\nvalid.",[14,126,127],{},"With a custom host:",[30,129,131],{"className":32,"code":130,"language":34,"meta":35,"style":35},"import { Ollama } from \"@arcforge\u002Fengines\"\n\nexport default defineAgent({\n    engine: Ollama({\n        model: \"qwen2.5:7b\",\n        host: \"http:\u002F\u002Flocalhost:11434\",  \u002F\u002F default\n    }),\n})\n",[37,132,133,147,151,161,169,179,195,201],{"__ignoreMap":35},[40,134,135,137,139,141,143,145],{"class":42,"line":43},[40,136,47],{"class":46},[40,138,51],{"class":50},[40,140,5],{"class":54},[40,142,57],{"class":50},[40,144,60],{"class":46},[40,146,64],{"class":63},[40,148,149],{"class":42,"line":67},[40,150,71],{"emptyLinePlaceholder":70},[40,152,153,155,157,159],{"class":42,"line":74},[40,154,77],{"class":46},[40,156,80],{"class":46},[40,158,84],{"class":83},[40,160,87],{"class":50},[40,162,163,165,167],{"class":42,"line":90},[40,164,93],{"class":54},[40,166,96],{"class":83},[40,168,87],{"class":50},[40,170,171,174,176],{"class":42,"line":111},[40,172,173],{"class":54},"        model:",[40,175,105],{"class":63},[40,177,178],{"class":50},",\n",[40,180,182,185,188,191],{"class":42,"line":181},6,[40,183,184],{"class":54},"        host:",[40,186,187],{"class":63}," \"http:\u002F\u002Flocalhost:11434\"",[40,189,190],{"class":50},",  ",[40,192,194],{"class":193},"sOLPB","\u002F\u002F default\n",[40,196,198],{"class":42,"line":197},7,[40,199,200],{"class":50},"    }),\n",[40,202,204],{"class":42,"line":203},8,[40,205,114],{"class":50},[25,207,209],{"id":208},"installing-models","Installing models",[14,211,212,213,216],{},"Open the model palette with ",[37,214,215],{},"*",". Ollama models are listed and filtered to your hardware,\nlabelled by fit — recommended, possible, or unlikely to run well. Select any model to\nstart downloading. Once complete it's available immediately.",[14,218,219],{},"Or pull directly:",[30,221,225],{"className":222,"code":223,"language":224,"meta":35,"style":35},"language-bash shiki shiki-themes dark-plus","ollama pull qwen2.5:7b\n","bash",[37,226,227],{"__ignoreMap":35},[40,228,229,231,234],{"class":42,"line":43},[40,230,12],{"class":83},[40,232,233],{"class":63}," pull",[40,235,236],{"class":63}," qwen2.5:7b\n",[25,238,240],{"id":239},"managing-running-models","Managing running models",[30,242,244],{"className":222,"code":243,"language":224,"meta":35,"style":35},":ollama ps        # show models currently loaded in VRAM\n:ollama unload    # unload a model from VRAM\n:ollama remove    # delete a model from disk\n:ollama update    # re-pull all installed models to latest\n:ollama stop      # stop the Ollama server\n",[37,245,246,257,267,277,287],{"__ignoreMap":35},[40,247,248,251,254],{"class":42,"line":43},[40,249,250],{"class":83},":ollama",[40,252,253],{"class":63}," ps",[40,255,256],{"class":193},"        # show models currently loaded in VRAM\n",[40,258,259,261,264],{"class":42,"line":67},[40,260,250],{"class":83},[40,262,263],{"class":63}," unload",[40,265,266],{"class":193},"    # unload a model from VRAM\n",[40,268,269,271,274],{"class":42,"line":74},[40,270,250],{"class":83},[40,272,273],{"class":63}," remove",[40,275,276],{"class":193},"    # delete a model from disk\n",[40,278,279,281,284],{"class":42,"line":90},[40,280,250],{"class":83},[40,282,283],{"class":63}," update",[40,285,286],{"class":193},"    # re-pull all installed models to latest\n",[40,288,289,291,294],{"class":42,"line":111},[40,290,250],{"class":83},[40,292,293],{"class":63}," stop",[40,295,296],{"class":193},"      # stop the Ollama server\n",[25,298,300],{"id":299},"switching-models-at-runtime","Switching models at runtime",[14,302,212,303,305],{},[37,304,215],{}," and select any pulled model. The change takes effect\nfor the next session.",[307,308,309],"style",{},"html pre.shiki code .sYbnZ, html code.shiki .sYbnZ{--shiki-default:#C586C0}html pre.shiki code .sTNBD, html code.shiki .sTNBD{--shiki-default:#D4D4D4}html pre.shiki code .s3F5K, html code.shiki .s3F5K{--shiki-default:#9CDCFE}html pre.shiki code .sKc5r, html code.shiki .sKc5r{--shiki-default:#CE9178}html pre.shiki code .sCudf, html code.shiki .sCudf{--shiki-default:#DCDCAA}html .default .shiki span {color: var(--shiki-default);background: var(--shiki-default-bg);font-style: var(--shiki-default-font-style);font-weight: var(--shiki-default-font-weight);text-decoration: var(--shiki-default-text-decoration);}html .shiki span {color: var(--shiki-default);background: var(--shiki-default-bg);font-style: var(--shiki-default-font-style);font-weight: var(--shiki-default-font-weight);text-decoration: var(--shiki-default-text-decoration);}html pre.shiki code .sOLPB, html code.shiki .sOLPB{--shiki-default:#6A9955}",{"title":35,"searchDepth":67,"depth":67,"links":311},[312,313,314,315],{"id":27,"depth":67,"text":28},{"id":208,"depth":67,"text":209},{"id":239,"depth":67,"text":240},{"id":299,"depth":67,"text":300},"Local inference on your machine. No API costs, no data leaving your machine. Requires\nOllama installed — Axon manages the server lifecycle automatically.","md",{},"\u002Fdocs\u002Fv2\u002Fagent\u002Fengines\u002Follama",{"title":5,"description":316},"docs\u002Fv2\u002Fagent\u002Fengines\u002Follama","fWf0ny2Lz96C8eIEZh4eDAVfz6TvBEhjN4XZ46TY_Gg",1785671806002]