From 21ec2f123dd1f8e2bbf6497cccb006d5210ed516 Mon Sep 17 00:00:00 2001 From: Shaivan Bhagat Date: Sat, 28 Mar 2026 17:15:08 +0100 Subject: [PATCH 1/7] include ollama manual --- docs/workspaces/imgs/ollama_pull-model.png | Bin 0 -> 20694 bytes docs/workspaces/utility/ollama-open-webui.qmd | 116 ++++++++++++++++++ 2 files changed, 116 insertions(+) create mode 100644 docs/workspaces/imgs/ollama_pull-model.png create mode 100644 docs/workspaces/utility/ollama-open-webui.qmd diff --git a/docs/workspaces/imgs/ollama_pull-model.png b/docs/workspaces/imgs/ollama_pull-model.png new file mode 100644 index 0000000000000000000000000000000000000000..7c4673ccaabf3be10f4cfad1eac7cc83cf3b350e GIT binary patch literal 20694 zcmdSBg&c{_w#9lSMFP-IQM6_HwG`>e;XVOtx7I4 z5Qi8zj}4BF_J00++iICwu1JxR$=;;u4$X%m#nYW{6+zmF->n|f+rKY5KBe+ErT@IG zyP%hd6{>4#J*NNScO({_@tsp6ubRGxFT8l$?d{TQsg_NbV+@6kyOSQ%WX(mz`SS7$S|>sHw?>^W|9R ziM(;UJoa%JTa1_v>$IG&HDB_i#B_9V`wl(VUjCt+m3a1O0wMj~Rg=|Z!8CG@7!eZWoqkBbB^s7%+RD>@!YQ+Z;mlRD6r+CSf-@fD!|^BLtmaz{CR1vNGBQ{URt*4EX_ z-p%``Zpt&zo3aiDM&3r68WPrSE_{|YZdSH@elG6FLr|prB)~@(TW?FapNq4rmxP}* z)1N&gz-Q!UekS;zUAnT#}b;0kV@ws28C0X_jH85}qqF6C)sC!wdP{P*GDH)$q& zZ*O-AetutHUp`-9J~z*Y{DR`*;`{|LSE?KKY!h)t)IhxS910G`?0_S@*}_D7vvM*|9?Bo+rjSt z<1pkme;xMcx&FGG6!K&eIu3re&c=!kE}&Jxt;q-qhzUsjxz7Lb)qi*T*P(`9ww?-Z zF5pOSng9Cb?}PvM%m3$rf39itKi3o&`oCTCf4=&kLy-?5q3!7aE@z2sMHxXU{{Lg! z-}_7PBb)etn*6W+{IeBws0@x2|G$Gn2B%w$o(Ba5fug1;XW)mrn~U9MI7QKa`{|Iu zP$n@xN17^?ux&Xu=E@_L0^!5!S&J}qB= zF9!bpda=0s{@Z!Lp5(?(?bzbQ<*4Mv<%x;Y+xMT@vYp>asYZmrIF6x}^`A{Fb02-? zJQjnZFv208G@~i0$GMjaHoh!fn`HeyxIdAvLyqnvnfM1{PSq^6k zE}w1W6Y7^}V5_I|3@ZQP(*O6?I9@?3pAQZ+Z@+0Y7dx4??pe!tKwoZHkjbR$3bXrv+)T-JsGRcUlY=0mw0xSb;AWWg1^x90B;zNczOQQvZeNOT(5 zoE@!Meu$?kREQ>6-hSJmwMM)~geDJH2lsA+ZA8Wmi9je54a6jUq<^wQv>0@;t5a(? zlBS01E2q*tB04`?pT4qKlYdt!=G^IgUC-V#m!cpzp=TlUIkY#!I#_v@n`bV4bZ&llT_(}JXj(w5y3 znRKbiX&l$@KYE%U`S{)VE?HTmxFc5Jr|RxFHz8%i46MC-fY zVMjkiYINGav~V%NoMiKI!2JE=NZ~OlVS_R~R`bREjvIPRd}>nH*4+j_6OWzgazpNb z)8%N(#WsISur2PMI2cVjWqPj;CR7-{;PGf7>g+~dkWuCe5`R?RJUGqdHE$EsUJ>%$ zTQ~qCrm7@Nxd%hdd8$Oy?|9^P4|{$9arwQst|hG<*XP`Cxt`}SLdNS5CRN~9zgV#{ zev>+mgJI%Gk!uycmQALqNyJ^&=7dhrYBUnQd(I{#B24`Z!Ixb4RS!DaQ^*H?bYU^j zG8kA21?<+(?0k*hEO=CSgBb74@fsCxcrbTUK%}r3d~-D!RMFBQ&y?H{7@f3jn{mJ4 z#gAthc+b~C@6i32F0t{K{bcvdbf>v+9$o$1n(7I|!0q`!<3`J_2P0I)C9ZdA zbo?~r4o0~zuAz!~){ODP+%_!?Wxr=DN+CC1!wQ2U+!*x{PE9_Vskr=Y0&TxOn|{8H z=!C|GkPohb_Ql_S|Jz(_DFnS6G^OjA%;^&Pg=(=&=k3VtP16qvGzD&uqh?H z?K8zZiq@8UfYn%BOSb{4Ss0hX_AcPNJt9ca`V+ki_8Tkkivk@~NQUll0|5 z&+0FC{95)Ti60!<(p0jC-$s>bTx{65?d~ZnC>gPUgqzs7pm&Yi5%_{<%q%D0QT|ZD5ab7Ib;oAJV{h-tCzv#MxRJR)SFk zaoV4)DNe~G(Gp`E7f;zMPS5pS{4(d!g+T#X86KZN2*J(yz%}$h&kgMWB2| zrp0^9%2fG-u-&kT)%whZ&rq7S3xy_`=dfV^?8&#_n_FnLXY1MXf+jG?)E7(|XV{|O zDn4^i-7csWB0Ri3!NHsHomnk{;yI5#8&6EljR``m7@<4i{k2?`?)VJqIleQb%RR%A zP_+#@g<>oRDnTp7&V~4c4nn!+wRF?&^X<28`;63OQ(BTEm){OFcmStL0pS}bneZVF+4OPWodD)LHcohDE(>S`@PR^Oyf~_1%*k`MXQyLd&KuFwIQTW0xm~^9bz6+Ml3~x{w^prUQs{M^|h|y-ql;G6yj%C6V z^cw8)5*swP97(zER@&zh<)Sj@fHFEg(Ea?Lxx>a8LO|lX?S}cLGaM%wySJV#esa^k zL@Sqi+mAc)$^LuadQ3r z=*BI1WBKylw(KWC3V}JBh$uqB+`IWQft9;=Gq0hc#LAsFW2Ozx4uL2yQSec2Y9XcZ zIFOv1n{WaN4m^6PSuIP6R{8a2uOU``=UcC7 zwk$A-gIMiVDC-0cx`v~JiQjEkwa%qW$w0%!7=PZvo75xHh$57|taiw5n!&&&@4Yf@ zqCbidfQQc(6s|2+wUlf7oqTzCZ{jjJ=}AG``pEBEHNtNJ#|gjOmLilXJlEGbW9poO znnh(pGZALbQYa=jvlY#c6m&NJG<%;Uvm2zrUj!0gBEWcPH5GfH9ROzXIS$AOWG*9& zrAC;WBY4Gsf85VFHTJd2c40BRoBxM_GMjfid!AoWeYP%R(Czpoor@xmW!t23BSw+T z_Dyf9GQ%n}|N3&OdPW!P) z@NX|zZ`F-6$j8mrc` zHgfyhyRR*ew!fq&NuV>aldu$Jg^a^5_B%RU^smB2DN~8MTO-q<*9$6emjv~9rs0|2!(iKJu@Y^{^J{ce{O#&L zF-}IYo-~qRT&n2k9v9bkF7%d;VnnYR^wPOb$TP-QZn`rD4F`GNP$*|-2Ers<-_m%h zeih?tQIvGv|NgZrYa%-dM_Sf&hTL-T88p9sirGlLxwgdt7nw!`nkF{U|;ex+18658&si>F+xyYI3{D8f+K0R3W8IHk``xy2< zr)6uF=+d;!H7B%8@<$GpORJMZ05*nlk@v#hJq3UJ06zkY5bY#Is0|J3FuT2#Al<6_ z#l?PbW^h(g`P7^I8?Wo6;{EW9wTY&$B0F2n@7C@It!F)aMo_)5u+m!>|1{=kbrSlhnH``laPL1~Ap^CX1mM9IdL|`NqZx=^q**IHzG2+-!L` z!q8KrmJ-ItE-OJr8G$Qj28me&2?Bv>Ep9Dac3xIqE*dM|_9C$4r=Wg~+F`o$)@bSU zOmeyt4x?C{=$wwagWAwty`I#~Wft|2?{*2xGk1@-<$$;0Qz%(Nb{gjlW3D-Z%MtfyY0e-e}+M zBY$O5SD0z+$A2HbZSS9qO~Urr^KRp5(^B;MZ(|+X%yPhKXuPWLPu%kl8iLC()WFwrAswRxEH5TrvN5@x7>_lj>(3iWb-O&vX$MP!D`YS1A z8pM(WUo0^Xf9#E>R*^#Mcy*+AOvG2NLs7y#7en5Ci)c2EQ9G6yGAsjL+J5;bx?di@ z?s_jUQlp(O^OF2X=;<) z5jKO6p^zNE`Q~5=(Gv3~8jw6}I<#6TvnB%0*wL~10+%g?kn50HC3_pHou#ozbgr8U zRkkoqAzUn8ermb_bp60Xik{&(PTS3KMq{jWW0i=or7I=tABm@DhQpPdF$5(}NrdRA z)&*f((bs*qYJT4PY1vkr*(ums=|p%Ltei^>vOl(tq(#z zut;pG)$>%b(vv{2x}WF6;;YnBrLDf(>7DMi?jLE5I>WMj_JfR>P; zX}mgKd2K37jjJ4V8rI~``t=={id4x^mDlaN#xYl-`)-1w3Fw?#0^Yqs8LzVtRh&AI1Gwb%{cJ5aH96 z%938%-m9i8ekUty!7CtS}?2tw;rhg=k9&wuBqb+ z6c|rbRH6(sKODz!@C^oIvY@`M9NAXeze)@o^7JWCb##fQfBh%C)G#Ld4BS{b6><@y zA$w-2gfn0z6DQAAEK(0U(e)0HN2G{g9)l#*xj-+11LCd|Lz&a0Wi0wbFAJO|O$I#Z zoo~XQ=&(a`ltOIfjmA`%kP4A11?pMT%eb5>5^$I;Wc*#VM4PJmP%>vRxFLTyJVBkl zlfxLCei+{QbgFW$&7v!8n6y+69Lg~TRbAC9Vd!u*huTn}itv(g-^JtE2XYEhxj_R@ zsZmfWguKR66mARkKLa335z)(y6jQ-D>LFIg3`CkJ7`S~{GMDln;R#{% zJ*JhI5Gi#fi-TJiihgwmJo%C=BxZ`~uM41q3pf{)3q>NRltW^gaQN;;-Jp7J%fmk+ zr5`E_GJePqHJV0n`XBA=T_kAJhGfHL`GR;RxQjYf;6aNEqmBmHmy~3@Tp5~11`$au zt(0>3`Wo4JptUf4KXF0s@B*dm)yc$<4-p70IDF+9Ns?T`f0CS@9C#kGv^14}MXmvC zNZo*ymK5eKdGI6#F#EYW2dVkqSd!I?IdJ0zJ8&H<>I~sgG3Eu|Z-FQCE{$J@2v%D3 zL4)8VAd={%yXzdLx~-F>$!|O~9_y~9>1$!X&JSuKs|C_aB$zR)N|ngB9;e^2pRKmM zYqdQ6_JOQ(({fZr{eF2x6%H1i_qgNM9T-}33?ddrp^7V;0od?)RReM?g&M*_a*(n~4I7TMe$tcHdxci2E(x?p(ck zgnU1^VbTzk5=K>AmdGUT?|s3I--23SrLvQ}$UuAX?AKJ1%(;E3ajl)$#ZIk4Dw`&n z2~324Pz1qi+JHyGrU3;yTA&fo+TbQNxaeI#Dv6w{cQT#CpfH0Gh<{IZ2TB2EHJVBo zMCST*H7K@0-XcLBUfheJcLc~qpC%$ROw1?t8TkOthHC3Z-V-VH!bYUlv2^c^Nxfm! z+S35pDXxt=_hump|e#=g-vuvgl>?xaklOM8a_cbNOIseF>dF z0TVW{1^Q#9lvUx8}r{9o&UUO0bFQAcKv$a}Ka9Kwkesw04+kbU@T=E3|v zXjQ_mt-gd0g3ew!h%G+eeiLE`VH98p%+1f1^7j-w-z?sx4fY#jmg$qoAiFqsHOQV1!88LZNgUKBz6qh26Jf!hc-`c|&fC_Mvl^))#)sokh$j(& zMei5#^%G0Bt`vXk&btFZCHRN4f4}~Uo=aX#S;t@cw(*Ny_NO`hS;)>rTH}N=4cUp_ z8U>n@6dH^;MlI+W?@G5yZcrrQWe=e&Eg(8{!f(Z5R}>yJd#+UjRUrb4Awb{~=%j_F zjjo0F-kMwX#t>Ql+?=okat>bO;^T^2H!E@c4X6Wv&$+#X2zm72-!?k4X*&Es<1&se zShQd8z$S7RsHpb!^BtQ-aoym-Xvuj^>gJk}lGgMa(-{r}G5Qp2p=bZTGH_UH4rEsS3L)WIClj>->P`Y98>OW*=t)dfkHk!4H zel^ui<|83&<~A$ zYqeuCaoIs|wPNLjBYg)KMjxb3Z~v`5u&pEV0AW$|4$KPfOQBd%tm^3zL09LI08mBh zYfFDpy(^M`5{Xa)(tU!u`R!+b*`}*ahW+?8CVNwtbW_#3_o9x|UcMS@3LB4BOu(Sr zKoE~wWMBZ5ne}(AWm=@;qpM#tNfr5ok~;!4U&I^=(e~SP-bgqcKBdYklu#bomfxs% z{&FPXy{SVpLK}3tTnH+eaLs~Tt*&K$;PtI;9X&wFpEN zQ?i3_%D0Ddb_HRhPsAg?T7`Axc!A+ZwM8lE<2>sxS2eFwGc$pb z?TXacnSLi_w}9o3q3iWCGNrF)1IrBXJb~anV4Z zYV~zf{}80D64ek*fOD(|RH*psrOk_(2f=+Rs)Lvf6Uo)S{aTng{AE zs+Lz5fr!q`0QATAhRY1T1UL$x_drbw@y|lVmEoCst=)J;d?V1*opKVxoG5TSb=c34 zpUL{o`dsi*AI=V~eQ7tELp9gnLW4<2@2u64%y_OeX-14?f!9`~)sd%d%BV95wHY=@ z^JDH)GHEkc#uDrI*_oZlS#wW*=O0mu%BJa(1T8}UR&sE7Bl7+6YS-9h5TMf`tET#NPr181aUV; zeOh-W(4=Ucaa4Gf0!m}~~G^_8J zYK+xn#f-z^@eP#z1$GHI>+p!Jr?D@iyf2NB2Gw|@+XDgxtA5Ma8rjSoca)!xw@TeN zQvr;rMnk+~pOP zmFqkR^e8i_FUP;8RoAfkIRz#FC#X$ilOo+Cl3V;=Fhh)GN=+C^`S#eul42^-1lbEQ z7wwD5bXf#3mOplX)JSFvLri$UG>LDXDHiH8^DAQ0JTg!StEFqOlsd*GmQYZ|EpBn^ zxcaQ@7i(7*HZ<Ugz!N7)X^5~U{;O0Ve zEfKN2P}3O=vC=m@OHW1tx+}I9JVVy*>p^09DyEFdxX4F4kcU@4s^iY`?&XEnfoYZ% z2-)B4(EZGZWll54Q3cLBJ<=(Nf9=Fm3G>j+x=J-;dcXzp?H1pB+XW{33GJL1=7RpxJC*HJC;)xQMkEwv@_AV^M=|R!& zk1EtIxV;!wJwcWcSt5M)w*nd%_UVPAC^He{0d^@Hwm-ZKHXllDX1^9D8IKGlwqpob zAJ|^ElTE^;^djzsJ`BRtn7~HujSp*LIWO92J03Mxoa8iWv36zTKK^?38V>2CAl8B*)RLSi;cdOLBy}eb2Lwy^K@M40ls%oda zIZC6l=zjH*XlGi}?RLv0rHl*q)2{Z}0~fY47bdz2kPQzP;6o_{yuoHf_8Wa zSjB3OB~t2GM$%N6Ojt$4Fx2FBxi=r@QlNcNws>O%d-eRmG}H8Y5O=_^;D}#HvyD>A z8FW|C1J`FBShjsR-`nrois(=K6l7Vh6JHBX=<0E;S#bdiFYi&}q65q=iincyz6h__ zkgjMy7|i&P1KibMLAuyyhc+Z4OZOO)byvD~n}gMZ8cmXw**csaX#rHdgX9HtW`^Z}GG!^jS9@7&NAw+# z_>}^XhrY8eXz;sog7KMW!{2G9~FCJDel|2FwO$pNp)2^4Y6{Qt{N z772<=1>poZ=$Lp=KmqGc9t2kJxVAl;=viz^XHF>I&~X=L+=!s(1b}09}N*6nu3jta~^LyNy2eoB?R1csYP9>WNp33(kqj81X1 z$-@uu{kwoS+meZS!$CLkF}VZb*`!vW--64t3)1uRAv;!{d5-DuJ?M+nQiPe<#)h~S z+jH)F1d@AA#aT}QT_c?8JsmE!IeX3q8InLVX8q%{F1G?(H=0o96poAo*wVRu?EoF` z0AFG&M0&JAkLDX(dTRRVod{acu?XT{Q~VwUBh}VWUUd8=++GXLlXHBW8FUbOc#M<= z9g;=oorTqllOiq)9F}tw!F&E4IalO(e4q+_U~ri^q37L8I#L`6YD5Hr2~DiI=sD$W zS2JL8TV0Bv7g{HZv(>qMlOd(>=(1TWVWdWzI0bo^ZeYO6Cm2ht)^d6Uav(0u=9WE`Umo{j=o$h|nH;kpMetW@PVVVCk85oR}Q6br}K-X(G92In- zWS=?zwLKqCf4h&x^eqVt2!2V%Z3e!%7d9x_t7itwg1chlNom75%|0C-|s-rmyrD z2!hmk<-FDeu4!KW=ez$(Mye#C@kA7t_mU0l2*jH?x%|5t;(4h06h5bQPp# zxc91k*Ph>t0=zW>SfM;f&K~NAgVY5sb{q50igP@NIF|v;$G`p6Ti*{9v5~uhXKNYy zX%6n0VQOg~krMY#Bw#WN-A_7zWohs9UOP$u%nFBGniovH=|WD;QpS9pD))_k#cf{Lt+__EZv+ z zBb+%=jER0RTX87Lh!Kpypm>8@)qz)krszMnG^qmSY1S(?(%n0kCU9zya;eaavpKAn zKAG)_Q5wsW=9Ll~4O1-Xz3IhxA`@36*OXa54v6*Cv!(CqK}qkI&L(={nlD6Zb|v@Q ztWB@+fCLucuuzt0VWedurY#ZoTJ4WgiJXr^i@>9=gr!F^d>q}_@Qzc)nkM8DoDPtC?kNHQvTk(nrfDt*w2E}Crx1VB5xJ2^+H zK`SEg%+O@}ms+ZK(n`PID(>W)WlU+HBrjuXSGkrR?L%ftx; zL9a*xI*CfXuYiU-u`$#7PWT{%j8dMrTt(YOhUlC&E@{kH-It8(_6IZL8@;MtqL^A_byk zqAI4B8V9Cb!#%?c8VbWq^3b2tO&+TV_a@-h2-5nRU5q1hV}S8SBv=!Pryz}WyjL=T z=d<;S^*>p+kSUe+s}s|yMWkbRmk)IxDoG&j`9?#e11XLFg?&PY8Ivc(_!fDfcSP<~ zy;g7*U{;DvNTnImpXE(#%#19Gdq$*>#*lVADmkVab$X|FrYymC?L7F$qKmxceOP%9;*a~zi)k!cR?jQ~C z90i1jnde?gsT^{imIKl}t4WfR@j~y)hLD;$M{2zWoGmk==$g%(!IajTfq8T^U$9Ua8F}dgSl`TIm2L<4Ol@ja@P{TawU9l8j4> zM@P@f4sOmqz4AiVup|KXu1_~BK@&8GB4UXt;0Wf_(u6^L`)M3Tod%$QbDEBs`f@`) z8$j;lG*y;$mVE^vhY^6BxxCUn6=<3kL`Z_J#asSmD^SD#L5vtkL&2ZeG5YVOMP<~1 z1K^1v{}S(?eSq441gYHG^1J^)PoOsgG5!@ICJlru|9;sA$bj(exu#Z1_uqk`NQoSp zBJ|Hb;NHW)b5u!T*DHWmL?|3_akA@zU#Ica9Yid8fKN)z*ZlpN2AT1u=cvrZjyn*8 zDb6MCgX_hC<1C5vgTR=-gNI5kL?`K!^9GH<1wzzhJnXG`=1%W5ELdasBPs4%w-UvD z8{F0pBsewtHC9Td948<>SzAAgL*LX7LA-FdJ%v)t95v&Gr(kY>(j?HZD?f;w6fXl zf*QA(nh#71y0B9K{xoadS7O=MO@{$kq6K0Y36hxq(YHYaXf5jonS2A8xd(m#x$xWo zImFE$2xr}XR6WszjrE_}lBIty2)6;XFFkmn$zvJlM1)!e%JGy$=fC&&ja?+c90H0` zhT$YgT2W5H6Y53k1k%Et+t1dQ%U1PmyUBM? zPIjnF8eQp-2{-}gS%HsUd(9hBfhS+s9j+5@FN%rSkHafL6}J37kkW9#y~glpTG#UY z=a#blfEN&GBarj4T=Q}iO*fD?dqPmrd2T>cQb;`dVYB82d}p7P5)i)n6w!btc*a^FT!_-~-C>cbmn0Pz19DEW=n8ia=D&@q$q50D(*+ zcLR?lz7C{@)=OT6T_UOgmSzy0wFvj~06Gi~0RO_>3QQ(%6bPk70)VK9oU2PFx}y6c zeLhv)pI62I^(?~)h)7~Ej(fDvO9_y58No)h`KjWb-*Tkxa-t;WnF@7^8XxmZn-jn5gc(v~@6L5L-Tg32`f;ubeD6PEbk$!xQ1y6v3Xm zrY>a43DjicM{qC)yP6$aW1lla-o=E7@EJqyDFH(>nGOQz+1pu_TDMZryqzqhWWy|C zm0CGsf_X@%?*K^h5!AO*o$Lor0k6Pf=e1YiK-cUCM8i_{_LaqAmbjNP@|%QBF#2(q z0p=pG*M<{hB`3+|kv0IY0>zR_%)P)t#jI9O=Pa2LQnZfNWeliO!EPCj2`>fv6MJI; zZpMX1+>)1TA2Tir8|2{?py1e$nUf*k#s*lY2GT2fp=J;Kn#6nIR1iSVf>#`N4wm^E zMlq`&J8Lufrpf!-SO$ z|9UdCF9=}w(Ug{JfYI?3BXsfo6B-XBg!tg+ z99HC)+J_rk`Kd;Hb~KQw=eRCiShhoHoKX~8NE(z+&>O1u2LOM3s-1*;EjKujls|J9 zYC{bCTpMD|=d7UhYM4SAR-bZLLB)oYF@z`DFZP@^V)iF%tW}t4n3bhyM$-4MpX_Uj zyz3NMy}jty6X0-aq;U7(rfIifJ|0^Ai|}@08`O?-5`MQX(#1T?AE!v+3q#;f-Oimb zH{22q4IQY~d)n7b(^r7{VW<*960cmpwq#K`e;?R*!{_ew*;%^>BcF`jtSs#Y*d!P+ zbZf_uH1IGt|3}F~rj%5~G z!i`2zlPvz$X)!{piGbJ(pzaqBaBDM9E!T^r2U(2gwZ+#+nRzri)+GXbK*6$Z7f+MR z9(r!dp5wosrCi>A{)3xaVy?vge%_v`otd1`sMu<}J2Hp|#7j}-XXGEpSaDvQ)JAoh z)NFYil@3w2aT!4?#s4hj5Y=bPC0CKn)4&)d>k0@}W2f>^txrhPeJ^2G!mu*hI^iTJ zXE_vm9&vhxTyjuHKRm>#nwc0MtTlbx7Le>(YZ2Nl=rzY>=KC}a`wNhmwQ|w$X)h2d zx##t5tKn_I(IayxkgO_tBLx-)0JY1j+;bn3{>`F!uLxA~z9fVL3twMSks~Q!?(dvoh2-o+OjVRcHK5RCy$++)>EjIZ-D0=~^nPDWTrF**8l6C_ z^YeblONWwa$8qj|uoU704A|X=pO*u&#mFRRUPkH6L-eR6yi0jim1A~F3tU3$sx7EO z%V6oxHYnds6{`@MCboO74JjLm1tV#cmF)0SNmwl!&m1M%owF5d#qYGD&Y4#+C2A3O zV*kZ?AipBy$~9!Jf6PxMB&VpJ9@OtlLgqWJRJ(En3ZUsY3zH)o32=_e^u)C^1cZq; z!fvVLf%zSG)*EmIVEI=x5&*yLh>cqvDOmp1?+82CK=}uZS~ZUsS_g(*&OMIh$#@h4 zc=FJ_Qdm26t#(o6@i&}zmB)f$3B{?y38N}<`3^8F3GcW1)F!VWv(rm60I=Y4({~^w z|2r>Oh{NT<$)~SNnsRfsA8ZLqaG_;S)S=CT5mNbM;Qbp(e2=J?6S&WcQ^MZuItJ!K zWv#oT&$>ws-#`AMd=C7N7pOi4NOfEF=*rJEzYE4n*f zHpR7u&={{?sslRIeiT3=YvJvZ+!c$AK)urp0kG&E*nYY@S}XJe%=)%AN)dfTQ$CtXKxNYm^mJ^5ahO7F za+<2EnYtGAh|oTf-3w_km@Ae7W*`87dg=M?d-svG3+&^*UB;8QsZ}$<(dtCt>{tfl z2m}*yTKmItNn*5DM4(t?OhVK4)SziT$dt?l$>%HV(8*9lS+=2p1by5dh`sz>JyfFr z-WM`RWBa?500zVYc!~I#(KN?@^@R#-8r0E!@!z0M1=suj-CDQ!pBES!Y6CX0C@B5U z7}WyO@3Yh-rG)>8&@(`kp5c+b0-+%tM(u1SC9s!<9q*e zvXiN_49eZ>1_JF3oj-uaLi6Ibn`=dX65v1e9~ppatpkMAYGlG48B>wHK4o^7_8ef0 zHu@TlMJT$WAPy?K+3D8n{JY%j^Vmow6s!Wv;AI0rKy@INze&hjp9k%RdLp32C!@kBx z0E3;Bj%rL=yp4_8{5(o322}m+*+3&@AApsTQ6QT!4ke8nTm(v2#q*@Qn?OA!P@(lu zTxy3Eu`z?!78j7M+j>^mHL~cKmE(^?{DJzD(3!Fie1e}m7_N-)0YE3{$ob6dFSH)- zz>i)dS8kY?17JN~q!f1mK#S5S`Tib&-L#;Z*zN`a;?WJ3!BMkmo{NDsa>7!-ru7>v z+zuO5>q7x(K9~K=x3Ku~Zta9fjwGzQbh|S&+zPi zZ(Dwpciuu)oma2UqY2*xFJ$oDC(aol^_J%DKTc7YsvectjsWCN*vxC>)-qT=RFiZB zI*o|=etwkWr}9X!#xWxJ`ivbc-0GK^uA^W&1mHLOL&W(pcTz*a8d6=W)Jkj)mTCGV z`ZcY`@FfU86b76<>H$l>h>#46nsqEB7LzF^$^j@gmLL}s?;tf~dX^}@Jq+rHDe9D( zfxYmU)y#dGNRZr!LW1^DFw}aIE>z&WT1M+*`3&zDPC!=UGUNnH9J$JCwq~_E6pJ1P zL{C9sszSr?S&clYF9V-aROqR+Kq#aq<)M;XAy*1og7i`SlTgU7%?ak8 zyYn+Ki_ciq<97}#i7|$MgQb<#_7g?p?SU6dyC4mdnT0V?o#Qj-unfF;%{K4)RiJzs zEj}gV{xGor>;Sw1ALcv~>r*%WAY_o!VMFR~l^_LlvIAh54|BAAOVjP^=zg${%y#46 zqT};^Vq`8sOMER&r{2+!`P=1NF|F^>JM>`vT-0Hk|1r|P=mF@|Md{U)Osha#@Jo+D zrY0^Eq_4+ZZPA6}g~1jDf?M-%0Z~1C%57L~`D4>bLpmKk8ppr%F`PFJ#pD;(#ApapP>1mi8 z06e?aIF^r(I|%7)n>1dCa4I{RM1cgzr_>cw8Ke6x-mw>Z2Q_(u)=_7B$n>7Fy?`1c zh~hkb&`&^}2N+&dnf~h!G6U)(4BtT%uU0C+FN6O9ebCL7uW)0HtlJ`v5D72jT@)vF zsEtP6WzFvKPL8TOy#^!M!tE^>>JR(Rm06a3O38fJ3?^WilAXZNYL7HX)Hre5F1_^v zxQp$VW>_0zvL-ukI_#q77=hFr<}a_2)Ia1th-sm}bekxORIWqj&syV8rba=UkME!( znv%1#f3tP zo`;I8yoMA?u4h`uBbF7PfUxD<4eUirDT0S+k6m}%{cvGii;M+ERJ{@ee`nDXc+>$S z)5r(PCbp1OFO=DoTST<%`G&36eyo_~diO%)fh<(lM+4rjEp^xge1vCa zdM*sT9P=PL{g1GtPyj5s8m*?bz|x{fq4OJnPo5>_6rVkHx&al15a@Qe;ns<*!%q*t zDBGi&U9R>E&Y@^1U+dbXB_KwV4ThK*7zRL%{1QW*it{2A=+N2v4%oYqbq8I&c*23_ z-WLPK_${Ea;M(4dtby zJ(kV$FXqY3ifZ+_W$7Dh$H5sfsrg>rn5gkMW6!pW_O;4)#Ts?R5ZE)0SM&jvE zJjTvpdJUdTywhb$DpuZWBHOgBXHOZzHb#2MvF@qjw$o=aYRR109CxWU?t~3lizL~uQ&2y{{LoPeST8N<)$lOLIB{!GA zPQE(^@}d;{b7~)egGi&T?u!*nDs!4FE26;{FwE!+)vCh4E}?{|*k2k6gp(=haq^#6 zGO9}v22`2(xn&Ce0uw>zXPnxU{*BQ16rv>Z2(ONQ1$%MoDOLD_RGU;Mv#0;z* zS%#;ROHY;)8&3n5X1b>xWSo-)?Y2YsOGpR8aaphv2G)`++(Z1`#G?N%Ru{aRC8dExDQR zzU>7^0?+S*aD#HX!^Ocev;6+z#yk&W9=DiW@8y>z=1AkiSoS<@eYdQmm)rUnPF{P}-{Z zKuqhj;h>|-EAl6!vq!^+3CbzTc3%6+%(W=ehCm(MeAliCGJkim~Y3^I1c ze35MJZs?Z0T@@!$K;Dp}vHmdaqe}M`y^ygFm#v_q(sq{$(_wMi(63m3Jx)Rub>(R> zcPqU(9J|~idnN%4%P{8jJ4M=xb@X+DVqK5Nn>U-Ccf)J5rQaE1c{sss#86%zO9^&_ zWJ_L=*XN!+&zPyjWET+~T*SV)kypsr2BmOYTXmNNtuF%Mz|7YN$u0+!t{{_Wps5U+ zO--PF=}x{ni_yyN5G3<>^ca~}`q!swgBh`lx>RbLv*TtH5G_a75_(|iqsyH%yvCmZPh{ra3cbroF5cISr9>(VmM6v=`zHkXoi$lH&_T#k4+Y9dfUq zt-kxNUk=;ZNpI;A437DKE}sSbVOJ^`+L!P-#w$(Ec%oCTa`X;W4m*SXwb?kEF8#)U z7l@;uak~6t4W^kO@6QO6#b{@a^2W839B5~jRWehU*p8%R>y0GtG5$xrqD^2G`Wv1wksA2i zLUYHL$u&jK2!vdo5lZ}X^;ir9bzG_7{WS68UQ>7h(aitV#JLAGal~;P;vq)@F%K++ zR!D+@G&}`cf~X`UfCPlfBfL=r>>)Z1P{ts0f!StY@T*Ki#tjGa_Wov6BH-cQ;wSw1qrx` zM?)L~Ai1nIqOcd$0LQOH!T2}D2eI8uaw#q2Ad4;oNr`T03HD{O(oFeK>5r+s6n2dU zUGS%3lxggrHAc(mCrh#m^`u; z(kU%#il1hx$NBidok$d!j>zWc(#M|DC4~OB+ZvL}h)gN%zX8FsN0Df*SQ*hu+8}K^ zQpQ)&McUcqU~lH*iP9_LQj)0aojlio;$E2i_%6LT~*>Y3F`0zk`j2Ud8)Y!&cS6> zUr0P9ipyO4roxAr5|<5>uf}xYat!c=!{PZ?sXGsZ2i@4N2g6+k80N!B(bu%L5vv}B zh;TTbFx_rxLglaXpO2tedg>`)x2Aq8Y5i1MyPH`YKx5>zASnsrcMAF|pJU#p7TCW4 zZz%>Gs5Z!~S%XS*yR#I&<2d0YlGyp!D@}dC?0G%Vj$?iaS@RmHdLl<=f+x0XE)>ch zJ1hKa&D3+gutj$e72??#>%Nxf_(U4iqVvGpAXu$;0d^HU-?hGKuV!KO^`J}qE0-DD zQ&#N989=+|1;e>G3sI?ugR+M?%=|es=^~+j@6R`{j2E6kXAga)Q%n|_47x8~3qXH# z1GS>#R>Z{WiCJx`=qM)l`T=N$(WWwmOqGOj)c6k(x-F#PPEl%6YtHfNie;~qd*)*s zv`!OJ9Gzl-G4R1CC#WAbbZ1G;>K>fm^dl>s_)S(MXZkJX`fn?wZsMl$yL-sc93-rU z7qLRKV!bCBjqpZJ;a(TMaSc7_oKoqF>>D|7?R$o1skIaAi2}Zei9PW>kn;9M57+4T zcR~lVo5&fSv}N8I;+2TgxRGiGw3=+d%`*exlX|EH{NmmC0H$Eesi@w)nQtzr_P8aVF+|wRkwq8H`U0q!Y+UvP3 zzeQ_ieOH(#ndB_XfV)^@?>1MQGsPM-w-?PG(*Ke69#qbF1mlnQ^FEPvgn!>!h41~H zD!~zsgocyXA<6Ea``1N|wgHtW?VSoNVip$Lva-5e8lPI1-7bmsW4@y@UwG(k921uJ z%^%WmXCU!LMX3T`$Z?IchZb?rU^HKwehZwPqY?zEgiR3_GDC<l7nmRD%@RaWf6FAS){ z&e;AMc9XNIPmU=NbHBxA0c+_2nGAs6&t-OW-Ws?LQEsW44rS#?Pe7oJX;Agvi9Kgg(U%MNja1|ONy9}Q0N{$AO zBFG`40YUUc`Rx44UuRC9;us-~BU%iz4G4 zv6wC9!KpZ(B0FhEUUe;Rm0ueXbnO4rmzHQ_M@QRa-3_<5_|tw-X9o@K*bSvJ7~|#9 zUXDEe1KsNJi b4t4(M-&WdNq{>OW5x9c8B parameters) + +**If you plan to download large models, create and attach a storage volume first.** + +See the [Getting started](../../first-steps.qmd#create-storage-volume) page for more info about how and why to create a storage volume. +::: + +Remember that you are responsible for making backups of any data yourself! Although data on a storage unit will not be deleted when you delete the workspace, it can still become corrupt. So if it is important to your work, make sure to make periodical (e.g. daily or weekly) backups of your projects. + +### Recommended Models and Resources + +Here are some popular Ollama models with their resource requirements: + +| Model | Size | CPU/GPU | Best For | +|-------|------|---------|----------| +| `llama3.2:1b` | ~1 GB | CPU | Testing, simple tasks, very fast responses | +| `llama3.2` | ~2 GB | CPU | General chat, quick responses | +| `mistral` | ~4 GB | CPU/ GPU | Balanced speed and quality | +| `llama3.1:8b` | ~5 GB | GPU recommended | High-quality responses, coding | +| `llama3.1:70b` | ~40 GB | GPU required | Best quality, complex reasoning | + +Find all available Ollama models here: [List of Ollama models](https://ollama.com/library) + +### Workspace Size Selection + +When creating your workspace, you'll need to select a processor configuration: + +**Since Ollama is resource-intensive, we recommend:** + +- **For small models (1B-8B):** 2-8 CPU cores or higher +- **For large models (13B+):** GPU workspaces + +Although larger models and GPUs are avaiable, you should not simply select the most hardware available (see [responsible use](../../responsible-use.qmd)). +If you are unsure, please [contact us](../../contact.qmd). + +::: {.callout-warning} +## Important: CUDA Component for GPU + +If you select a workspace with a GPU, select the **CUDA box** under Optional Components in the next step during workspace creation. Without CUDA, the GPU will not be available to the workspace. +::: + +{{< include ../partials/_creation.qmd >}} + +## Access + +This workspace can be accessed via the yellow 'Access' button, or by opening the URL listed in the dashboard in your browser. Any member of the collaboration can login to the workspace [using Single-Sign On](../../first-steps.qmd#webapplications-that-support-single-sign-on.). + +## Usage + +### Pulling Models from Ollama + +When you access the workspace for the first time, a default model `smollm:135m` is already loaded. However, you may want to download and use other models. +See [Ollama models](https://ollama.com/library) for a complete list of models. + +To use a different model, simply copy the name of the desired model and select the model dropdown in the top left. From there, you can search for the required model and pull it. +Pulling a model may take some time depending on the size of the model. Once finished, you can use the pulled model on your workspace. + +![](../imgs/ollama_pull-model.png) + +::: {.callout-note} +Model downloads are stored on your workspace. If you attached a storage volume, they'll be preserved even if you delete and recreate the workspace. + +**NOTE: Cloud-only models CANNOT be pulled**. +::: + +### Getting Started + +Once you have pulled a model, you can begin using the features available such as **Chatting with models** or **Uploading documents** to analyze files. + +For detailed guides on using Open WebUI's features, see the [Open WebUI documentation](https://docs.openwebui.com/) and [Ollama documentation](https://github.com/ollama/ollama/tree/main/docs). + + +::: {.callout-tip} +If you have pulled several models but are not using them anymore, then remove them. Go to Settings → Admin Settings in the bottom → Models, and then disable the models you no longer use. + +Keep your storage volume in mind as large models and tasks can take up a lot of space. + +Keep the costs of running the workspace in mind as well, and pause the workspace when not using it. Refer to the [cost calculator](../../responsible-use.qmd#cost-calculator) for an estimate of the costs. +::: + +Review [Utrecht University's AI policy and guidelines here](https://www.uu.nl/en/organisation/ai-policy). From aac020199ef0fdb70942ae561611358e9e7f7064 Mon Sep 17 00:00:00 2001 From: Shaivan Bhagat Date: Wed, 1 Apr 2026 20:35:10 +0200 Subject: [PATCH 2/7] Apply suggestions Co-authored-by: Jelle Treep <40466121+jelletreep@users.noreply.github.com> --- docs/workspaces/utility/ollama-open-webui.qmd | 11 ++++++----- 1 file changed, 6 insertions(+), 5 deletions(-) diff --git a/docs/workspaces/utility/ollama-open-webui.qmd b/docs/workspaces/utility/ollama-open-webui.qmd index b370c87..e26b1b8 100644 --- a/docs/workspaces/utility/ollama-open-webui.qmd +++ b/docs/workspaces/utility/ollama-open-webui.qmd @@ -40,7 +40,7 @@ Before creating your workspace, consider: See the [Getting started](../../first-steps.qmd#create-storage-volume) page for more info about how and why to create a storage volume. ::: -Remember that you are responsible for making backups of any data yourself! Although data on a storage unit will not be deleted when you delete the workspace, it can still become corrupt. So if it is important to your work, make sure to make periodical (e.g. daily or weekly) backups of your projects. +Remember that you are responsible for making backups of any data yourself! Although data on a storage unit will not be deleted when you delete the workspace, storage volumes are not backed up. Make sure to regularly backup your files, especially files or edits to code that are not easy to recreate when lost. ### Recommended Models and Resources @@ -62,10 +62,11 @@ When creating your workspace, you'll need to select a processor configuration: **Since Ollama is resource-intensive, we recommend:** -- **For small models (1B-8B):** 2-8 CPU cores or higher -- **For large models (13B+):** GPU workspaces +- **For small models (1B-8B):** 2-16 CPU cores, depending on size +- **For large models (10B+):** GPU workspaces -Although larger models and GPUs are avaiable, you should not simply select the most hardware available (see [responsible use](../../responsible-use.qmd)). +These are rough estimations, it may help to check the requirements of the model you would like to use online. +Additionally, to spend energy and credits wisely, it is useful to experiment a bit with different (small to large) workspace configurations to find the appropriate configuration where your pipeline runs satisfactory (see [responsible use](../../responsible-use.qmd)). If you are unsure, please [contact us](../../contact.qmd). ::: {.callout-warning} @@ -95,7 +96,7 @@ Pulling a model may take some time depending on the size of the model. Once fini ::: {.callout-note} Model downloads are stored on your workspace. If you attached a storage volume, they'll be preserved even if you delete and recreate the workspace. -**NOTE: Cloud-only models CANNOT be pulled**. +**NOTE: Cloud-only models CANNOT be pulled, and are not recommended when data cannot be send to online services**. ::: ### Getting Started From 6c0d04e118ed96e69c807a9d333cf76321764fa4 Mon Sep 17 00:00:00 2001 From: Shaivan Bhagat Date: Thu, 2 Apr 2026 13:38:02 +0200 Subject: [PATCH 3/7] update ollama --- docs/workspaces/utility/ollama-open-webui.qmd | 63 ++++++++++++++++--- 1 file changed, 56 insertions(+), 7 deletions(-) diff --git a/docs/workspaces/utility/ollama-open-webui.qmd b/docs/workspaces/utility/ollama-open-webui.qmd index e26b1b8..4e685bd 100644 --- a/docs/workspaces/utility/ollama-open-webui.qmd +++ b/docs/workspaces/utility/ollama-open-webui.qmd @@ -46,13 +46,17 @@ Remember that you are responsible for making backups of any data yourself! Altho Here are some popular Ollama models with their resource requirements: -| Model | Size | CPU/GPU | Best For | -|-------|------|---------|----------| -| `llama3.2:1b` | ~1 GB | CPU | Testing, simple tasks, very fast responses | -| `llama3.2` | ~2 GB | CPU | General chat, quick responses | -| `mistral` | ~4 GB | CPU/ GPU | Balanced speed and quality | -| `llama3.1:8b` | ~5 GB | GPU recommended | High-quality responses, coding | -| `llama3.1:70b` | ~40 GB | GPU required | Best quality, complex reasoning | +| Model | Size | Parameters | CPU/GPU | Best For | +|-------|-----|-----|------|----------| +| `qwen2.5:0.5b` | ~0.4 GB | 0.5B | CPU | Very fast responses, testing, simple coding | +| `llama3.2:1b` | ~1 GB | 1B | CPU | Multilingual knowledge retrieval, simple tasks| +| `gemma2:2b` | ~1.6 GB | 2B | CPU | Text generation, natural language processing research| +| `qwen2.5:3b` | ~2 GB | 3B | CPU | Coding and mathematics, data analysis | +| `llama3.2:3b` | ~2 GB | 3B | CPU | Summarisation, prompt rewriting, tool use | +| `mistral:7b` | ~4 GB | 7B | CPU/GPU | Reasoning, text generation, comprehension| +| `llama3.1:8b` | ~5 GB | 8B | GPU recommended | Coding, complex tasks | +| `qwen2.5:14b` | ~9 GB | 14B | GPU recommended | Advanced reasoning and coding | +| `gpt-oss:20b` | ~12 GB | 20B | GPU (A10) | Powerful reasoning, agentic tasks, fine-tunable| Find all available Ollama models here: [List of Ollama models](https://ollama.com/library) @@ -105,6 +109,51 @@ Once you have pulled a model, you can begin using the features available such as For detailed guides on using Open WebUI's features, see the [Open WebUI documentation](https://docs.openwebui.com/) and [Ollama documentation](https://github.com/ollama/ollama/tree/main/docs). +A good way to get started is to test the model's capabilities by asking it questions or giving it tasks related to your research. You can also experiment with uploading documents and asking the model to analyze them. + +Here you can find some [practical usage tips for managing and interacting with language models](https://open-webui.com/how-to-use-open-webui/). +This notebook provides a detailed example of how to [parse scientific references, extract quotes from news articles and work with a larger dataset](https://bit.ly/extractive-ai). + +### Using Tools + +Open WebUI supports **Tools** that extend model capabilities beyond chatting. Tools allow models to perform actions like web searches, calculations, code execution, and interact with external APIs. + +**Common uses:** + +- Enable web search for current information beyond training data +- Run Python code for data analysis or visualizations +- Create custom tools to integrate with your APIs or databases +- Generate images or perform specialized tasks + +**Learn more:** + +- [Open WebUI Tools Documentation](https://docs.openwebui.com/features/extensibility/plugin/tools/): Learn how to create and use tools +- [Available Tools](https://openwebui.com/tools/): Browse community-created tools + +### Using Functions + +Open WebUI supports **Functions** that allow you to customize and extend how models behave and process information. Unlike tools (which perform external actions), functions modify the model's input/output pipeline. + +**Common uses:** + +- Pre-process user messages before sending to the model +- Post-process model responses for custom formatting +- Add custom logic for specific workflows +- Filter or transform content automatically + +**Learn more:** + +- [Open WebUI Functions Documentation](https://docs.openwebui.com/features/extensibility/plugin/functions/): Learn how to create and use functions +- [Available Functions](https://openwebui.com/functions/): Browse community-created functions + +#### Python Integration + +For advanced automation and batch processing, you can also use the **Ollama Python library** to integrate models into your workflows programmatically. +See [Ollama Python Library](https://github.com/ollama/ollama-python) for installation instructions and [Python Examples](https://github.com/ollama/ollama-python/tree/main/examples) for usage examples. + +To use the Python library with your Open WebUI workspace, the API must be enabled. See [Open WebUI API Keys](https://docs.openwebui.com/reference/monitoring/) for instructions on enabling API access. + + ::: {.callout-tip} If you have pulled several models but are not using them anymore, then remove them. Go to Settings → Admin Settings in the bottom → Models, and then disable the models you no longer use. From 4d19b56507020bacb745d8545e1e475657a87a2b Mon Sep 17 00:00:00 2001 From: Shaivan Bhagat Date: Sat, 11 Apr 2026 15:37:14 +0200 Subject: [PATCH 4/7] include suggestions --- docs/workspaces/utility/ollama-open-webui.qmd | 11 +++++------ 1 file changed, 5 insertions(+), 6 deletions(-) diff --git a/docs/workspaces/utility/ollama-open-webui.qmd b/docs/workspaces/utility/ollama-open-webui.qmd index 4e685bd..7e13994 100644 --- a/docs/workspaces/utility/ollama-open-webui.qmd +++ b/docs/workspaces/utility/ollama-open-webui.qmd @@ -112,7 +112,6 @@ For detailed guides on using Open WebUI's features, see the [Open WebUI document A good way to get started is to test the model's capabilities by asking it questions or giving it tasks related to your research. You can also experiment with uploading documents and asking the model to analyze them. Here you can find some [practical usage tips for managing and interacting with language models](https://open-webui.com/how-to-use-open-webui/). -This notebook provides a detailed example of how to [parse scientific references, extract quotes from news articles and work with a larger dataset](https://bit.ly/extractive-ai). ### Using Tools @@ -146,16 +145,16 @@ Open WebUI supports **Functions** that allow you to customize and extend how mod - [Open WebUI Functions Documentation](https://docs.openwebui.com/features/extensibility/plugin/functions/): Learn how to create and use functions - [Available Functions](https://openwebui.com/functions/): Browse community-created functions -#### Python Integration - -For advanced automation and batch processing, you can also use the **Ollama Python library** to integrate models into your workflows programmatically. -See [Ollama Python Library](https://github.com/ollama/ollama-python) for installation instructions and [Python Examples](https://github.com/ollama/ollama-python/tree/main/examples) for usage examples. +::: {.callout-tip} +## Using Ollama Models from Python -To use the Python library with your Open WebUI workspace, the API must be enabled. See [Open WebUI API Keys](https://docs.openwebui.com/reference/monitoring/) for instructions on enabling API access. +Do you want to use Ollama models from within Python code? Check out the [Ollama from Python manual](../../manuals/ollama.qmd) for instructions on setting up Ollama in Python workbench and UU VRE workspaces. +::: ::: {.callout-tip} +# General Tips for Responsible Use If you have pulled several models but are not using them anymore, then remove them. Go to Settings → Admin Settings in the bottom → Models, and then disable the models you no longer use. Keep your storage volume in mind as large models and tasks can take up a lot of space. From a54ddf7e0b57a11e50a284089af97a66ea90e9cf Mon Sep 17 00:00:00 2001 From: Shaivan Bhagat Date: Thu, 16 Apr 2026 21:22:29 +0200 Subject: [PATCH 5/7] include api access --- docs/workspaces/utility/ollama-open-webui.qmd | 32 +++++++++++++++++++ 1 file changed, 32 insertions(+) diff --git a/docs/workspaces/utility/ollama-open-webui.qmd b/docs/workspaces/utility/ollama-open-webui.qmd index 7e13994..61eb1aa 100644 --- a/docs/workspaces/utility/ollama-open-webui.qmd +++ b/docs/workspaces/utility/ollama-open-webui.qmd @@ -151,6 +151,38 @@ Open WebUI supports **Functions** that allow you to customize and extend how mod Do you want to use Ollama models from within Python code? Check out the [Ollama from Python manual](../../manuals/ollama.qmd) for instructions on setting up Ollama in Python workbench and UU VRE workspaces. ::: +## Using API Keys + +If you want to interact with your Ollama models programmatically from scripts or applications instead of only using the web interface, you can enable API access. + +You can use the [Open WebUI API keys](https://docs.openwebui.com/features/authentication-access/api-keys) to: + +- Automate model interactions from scripts +- Integrate Ollama into data processing pipelines +- Batch process multiple documents +- Build custom applications using your models + +### Enabling API Access + +Open WebUI API is disabled by default, however you can enable it during the last step of creating your workspace. + +Under Workspace Parameters, you can enable API access by entering **`yes`** in the **"Expose OpenWebUI API (yes/no)"** parameter + +::: {.callout-tip} +Only enable the API if you need programmatic access. Exposing Open WebUI API bypasses SRAM authentication and uses API keys for security instead. +::: + +### Setting Up and Using API Keys + +To see how to create and set up an API Key, follow this manual: [Open WebUI API Keys Manual](https://docs.openwebui.com/features/authentication-access/api-keys#getting-started) + +**API Endpoint:** The API is accessible at `https://your-workspace-url/ext/api` (Note: uses `/ext/api`, not `/api`) + +Additional documentation on Open WebUI API access can be found here: [Open WebUI API Documentation](https://github.com/UtrechtUniversity/src-component-openwebui?tab=readme-ov-file#api) + +::: {.callout-important} +Do not share your API keys with anyone, and do not expose them in public code repositories. Treat them like passwords to prevent unauthorized access to your models and data. +::: ::: {.callout-tip} From 280a8c5922de574a9570e62f3b1bc7c53e6ccbc1 Mon Sep 17 00:00:00 2001 From: Shaivan Bhagat Date: Thu, 23 Apr 2026 13:17:35 +0200 Subject: [PATCH 6/7] include suggestion Co-authored-by: Dawa Ometto --- docs/workspaces/utility/ollama-open-webui.qmd | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/docs/workspaces/utility/ollama-open-webui.qmd b/docs/workspaces/utility/ollama-open-webui.qmd index 61eb1aa..3657ef9 100644 --- a/docs/workspaces/utility/ollama-open-webui.qmd +++ b/docs/workspaces/utility/ollama-open-webui.qmd @@ -164,7 +164,9 @@ You can use the [Open WebUI API keys](https://docs.openwebui.com/features/authen ### Enabling API Access -Open WebUI API is disabled by default, however you can enable it during the last step of creating your workspace. +Open WebUI API is disabled by default. This is to be as secure as possible by default, as using the API means that SRAM authentication must be disabled for the API route. + +However, you can enable the API during the last step of creating your workspace. Under Workspace Parameters, you can enable API access by entering **`yes`** in the **"Expose OpenWebUI API (yes/no)"** parameter From 39dc585b7f4a7c902db7f029cdcccc0aaa1bfc5e Mon Sep 17 00:00:00 2001 From: Shaivan Bhagat Date: Thu, 23 Apr 2026 13:27:22 +0200 Subject: [PATCH 7/7] include http --- docs/workspaces/utility/ollama-open-webui.qmd | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/docs/workspaces/utility/ollama-open-webui.qmd b/docs/workspaces/utility/ollama-open-webui.qmd index 3657ef9..3d67e01 100644 --- a/docs/workspaces/utility/ollama-open-webui.qmd +++ b/docs/workspaces/utility/ollama-open-webui.qmd @@ -182,6 +182,13 @@ To see how to create and set up an API Key, follow this manual: [Open WebUI API Additional documentation on Open WebUI API access can be found here: [Open WebUI API Documentation](https://github.com/UtrechtUniversity/src-component-openwebui?tab=readme-ov-file#api) +::: {.callout-note collapse="true"} +## For Advanced Users: Additional HTTP Basic Authentication +For an extra layer of security, you can add HTTP Basic Authentication to the API endpoint in addition to API keys. + +To see how to set up HTTP Basic Authentication refer to the [Additional API Security documentation](https://github.com/UtrechtUniversity/src-component-openwebui?tab=readme-ov-file#additional-api-security) +::: + ::: {.callout-important} Do not share your API keys with anyone, and do not expose them in public code repositories. Treat them like passwords to prevent unauthorized access to your models and data. :::