o § iõã6@sl±UdZddlZddlmZddlTdZeddd d ged d edd ed e dddde dddde dddde dddde dddde ddddgdddge de e ddedd gdddd e ed!ƒed"ƒd#d$e d%e e d&de d'dgdd(d)d e ed*ƒed+ƒed,ƒed-ƒd.d$e d/e e d/de d0dgdd1d2d e ed3ƒed4ƒed5ƒed6ƒd.d$e d7e e d8de d9de d:de d;dgdde d>dd?d)d e ed@ƒed5ƒedAƒedBƒd.d$e dCe dDddEdFd e ed6ƒedGƒedHƒedIƒd.d$e dJe dJddKdLd e ed3ƒed4ƒed5ƒed6ƒd.d$e dMe e dMde dNde dOde dPdgddQdRd e ed+ƒedBƒedSƒedTƒd.d$e dUedUdVdWe edXƒedYƒd#dZe d[ed[dVdWe edXƒed\ƒd#dZe d]e e d]de d^ded_dVed`dVgddadbd e ed6ƒedGƒedHƒedIƒd.d$e dce e dcde dddgddedbd e ed6ƒedGƒedHƒedIƒd.d$e dfe e dgde dfde dhdedidVe djdgddkdld e ed3ƒed4ƒed5ƒed6ƒd.d$e dme e dmde dndgddodpdqe eed3ƒed edrƒdsgdteed4ƒed eduƒdsgdteed5ƒed edvƒdsgdteed6ƒed edwƒdsgdtd.d$e dxedxdVdWe ed!ƒed"ƒd#dZgdyedzd{d|d}ge d~ed~dVe edHƒedHƒd#de d€ed€dVe edƒedƒd#de d‚ed‚dVe edSƒedSƒd#de dƒedƒdVe edƒedƒd#dgd„ed…d†d‡dˆged‰d ed e dŠddde d‹dddgdddge dŒedd e ed5ƒedvƒd#de dŽe edŽd gddde ed‘ƒed’ƒed“ƒd”d•e d–e ed–d gdd—d˜e ed™ƒedšƒed›ƒd”d•e dœe edœd gdddže ed*ƒedŸƒed ƒd”d•gd¡ed¢d£d¤d¥gd¦ge d§e ed§dVed¨dVed©dVedªdVgde edSƒd«d‘e d¬e ed¬dVed­dVgde ed®ƒd«d‘e d¯e ed¯dVed°dVed±dVgde ed²ƒd«d‘e d³e ed³dVed´dVedµdVed¶dVgde ed²ƒd«d‘e d·e ed·dVed¸dVed¹dVedºdVed»dVgde edHƒed²ƒd#d‘e d¼ed¼dVe ed½ƒedHƒd#d‘e d¾ed¾dVe ed+ƒed²ƒd#d‘e d¿ed¿dVe ed3ƒed-ƒd#d‘e dÀedÀdVe ed3ƒed-ƒd#d‘e dÁe edÁdVedÂdVedÃdVgde edHƒed²ƒd#d‘e dÄe edÄdVedÅdVgde ed3ƒed-ƒd#d‘e dÆe edÆdVedÇdVedÈdVgde edɃedʃd#d‘e dËedËdVe ed̃edɃd#d‘e dÍedÍdVe ed̃edɃd#d‘e dÎedÎdVe ed̃edɃd#d‘e dÏe edÏdVedÐdVedÑdVgde edʃedÒƒd#d‘e dÓedÓdVe ed̃edɃd#d‘e dÔe edÔdVedÕdVedÖdVed×dVgde ed̃edɃd#d‘e dØe edØdVedÙdVgde ed̃edɃd#d‘e dÚe edÚdVedÛdVgde ed²ƒed½ƒed!ƒd”d‘e dÜe edÜdVedÝdVgde ed®ƒedSƒedÞƒd”d‘e dße edßdVedàdVgde edSƒedAƒed®ƒd”d‘e dáe edâdVedádVgde edTƒed6ƒd#d‘e dãedãdVe edäƒedBƒed̃d”d‘e dåe edådVedædVgde edäƒedBƒed̃d”d‘e dçe edçdVedèdVe dédgddêe edëƒedìƒedvƒed̃edíƒdîdï‘e dðe dðde edvƒed5ƒedñƒed̃ed5ƒedíƒdòd‘e dóe dóde edTƒedäƒedíƒedôƒedäƒedõƒdòd‘e döe edödVed÷dVgddødùdúe edBƒedûƒed̃d”d$‘e düe edüdVedýdVgddþdÿdúe ed@ƒedƒed²ƒd”d$‘e de eddVeddVgddddúe edƒedƒed®ƒd”d$‘e deddVe edHƒed²ƒd#d‘e deddVe ed̃edɃd#d‘e d ed dVd d e ƒd•‘e d e ed dVed dVeddVeddVgde ed6ƒeduƒedʃd”d‘e de eddVeddVgde edƒedƒedƒd”d‘e de eddVeddVgde ed²ƒed½ƒed!ƒd”d‘e de eddVeddVgde edƒedƒedƒd”d‘e de edd edd gde edƒedƒedƒd”d‘e deddVdde ed+ƒed+ƒd#d•‘e deddVd d!e edSƒedSƒd#d•‘e d"ed"dVd#d$e edSƒedSƒd#d•‘e d%ed%dVd&d'e ed(ƒed›ƒd#d•‘e d)ed)dVd*d+e edƒedSƒd#d•‘e d,ed,dVd-d.e ed(ƒed/ƒd#d•‘e d0ed0dVd1d.e ƒd•‘e d2ed2dVd3d4e ƒd•‘e d5ed5dVe ed6ƒd«d‘e d7ed7dVe ed8ƒd«d‘e d9ed9dVd:d;e ed<ƒed<ƒd#d•‘d=ed>d?d@dAdBged>d ed>d ed e dCddde dDdddgdEddge dFe edFdVe dGde dHdgddIdJdKdLe ed/ƒedMƒd#dNe dOe edOdVe dPde dQdgddRdSdêdTe edUƒedVƒd#dNe dWe edWdVe dXde dYdgddZd[d\d]e edSƒedSƒd#dNe d^e ed^dVe d_de d`dgddadbdKdLe ed®ƒed*ƒd#dNgdyedcdddedfge dgdedcd ed dhge dddde ddddgdddge diedidVdjdke ed+ƒed²ƒd#d•e dle dlddmdne edäƒed̃d#d•e doe edodVedpdVgddqdre edëƒedvƒd#d•e dse edsdVedtdVgddudve edäƒed̃d#d•e dwe edwdVedxdVgddydze ed{ƒedëƒd#d•gdyed|d}d~dgd€ed|d ed e dCddde dd‚gddde dƒd„gd…dde dDdddgdEddge d†e e d†ded‡dVgddˆd‰dŠee ed‹ƒed™ƒedƒd”dŒeeejddejjdŽejddejjdŽde ed‘ƒed(ƒedƒd”d’gd$e d“e ed“dVe d”ded•dVgdd–d—dŠee ed‹ƒed™ƒedƒd”dŒeeejddejjdŽejddejjdŽde edƒed›ƒed˜ƒd”d’gd$gd™edšd›dœdge džded e dCddde dd‚gddde dƒd„gd…dde dDdddgdEddge d•edŸdVd d¡d¢e ed3ƒed!ƒd#d$e d£ed¤dVd¥d¦d¢d§e ed¨ƒed¨ƒd#dNe d©edªdVd«d¬d§e edSƒedSƒd#d­e d®ed¯dVd°d±d¬d§e edŸƒedŸƒd#dNe d²ed³dVd´d±dqe edµƒed¶ƒd#d$e d·ed¸dVd¹dºdêd§e ed¨ƒed¨ƒd#dNe d»ed¼dVd½d¾dêe edµƒed¶ƒd#d$gd¿edÀdÁdÂdÃdÄgedÅd e edÀd edÆd edÅd gdedÇe dÈddde dÉedÊdËedÌdVdÍdÎgddde dÉedÊdËedÏdVdÍdÎgdÐdde dÑedÊdËedÏdVdÍdÎgdÒdde dÓddde dÔdddgddÕded e dddde dddde dddde dddde dddde ddddgdddge d&ed&d d e ed*ƒed+ƒed,ƒed-ƒd.dïe d/ed/d d e ed3ƒed4ƒed5ƒed6ƒd.dïe d8ed8d d e ed3ƒed4ƒed5ƒed6ƒd.dïe d>ed>d d e ed@ƒed5ƒedAƒedBƒd.dïe dDedDd e ed6ƒedGƒedHƒedIƒd.de d^e ed^d ed_d gdd e ed6ƒedGƒedHƒedIƒd.dïe dje edjd edid gdd e ed3ƒed4ƒed5ƒed6ƒd.dïe dÖedÖdVd×dØd\dÙe edûƒedÚƒd#dNe dÛedÛd dÜdÝdqe eedìƒedêedëƒdsgdteedÞƒedêed{ƒdsgdteed5ƒedêedvƒdsgdtd”d$e dßedßd dàdádqe eedBƒedêedäƒdsgdteedTƒedêed̃dsgdtd#d$e dâe edâd edãdVedädVedådVgddædçdqe edSƒeedƒedqedèƒdsgdted®ƒedéƒdêd$e dëedëd dìdídqe edìƒed5ƒd#d$e dîe edîdVedïdVedðdVgddñdòe ed5ƒedAƒedäƒed+ƒedSƒdód•e dôe edôdVedõdVgddöd÷dqdøe ed5ƒedɃd#dNe dùe edùdVe dúdgddûdüdqe edSƒedýƒed®ƒed5ƒedAƒdód$e dþe edÿd edd eddVedþdVeddVgdddde edëƒedvƒd#de de dddd d e eedBƒed edäƒdsgdteedûƒed ed@ƒdsgdteed̃ed ed6ƒdsgdtd”de d ed dVe edëƒd«de d ed dVd dde eedìƒedêedëƒdsgdteedÞƒedêed{ƒdsgdteed5ƒedêedvƒdsgdtd”de deddVdÜdÝdqde eed{ƒedêedìƒdsgdteedýƒedêed8ƒdsgdteedëƒedêed5ƒdsgdtd”dNe de e dde ddgde ed½ƒed²ƒed3ƒedƒdde de eddVeddVgdddØd\de edûƒedÚƒd#dNe deddVdddde eedBƒedêedäƒdsgdteedƒedêed ƒdsgdteedTƒedêed̃dsgdtd”dNe d!e e d"ded!dVgdd#d$e ƒd•e d%e e d%dgdd&d'e ƒd•gdyed(d)d*d+ged e dCddde dDdddgdddge d,ed,dVd-dKe edMƒed.ƒd#d/e d0ed0dVe ed(ƒed(ƒd#de d1e ed1dVed2dVgdd3e edŸƒedŸƒd#d4e d5ed5dVe ed6ƒed7ƒd#de d8ed8dVe ed6ƒed7ƒd#de d9ed9dVd:e edƒed,ƒd#d4e d;ed;dVe ed<ƒed<ƒd#de d=ed=dVe ed<ƒed<ƒd#de d>ed>dVe ed?ƒed?ƒd#de d@ed@dVe ed‘ƒed‘ƒd#de dAedAdVe ed¨ƒed¨ƒd#de dBedBdVe ed¨ƒed¨ƒd#de dCedCdVe ed6ƒed.ƒd#de dDedDdVdEe ed6ƒed7ƒd#d4e dFedFdVe edŸƒedŸƒd#de dGedGdVe edéƒed*ƒd#de dHedHdVe ed6ƒed7ƒd#de dIedIdVe edƒed,ƒd#de dJedJdVe edKƒedKƒd#de dLedLdVe edMƒedMƒd#de dNedNdVdOdKe edŸƒedvƒd#d/e dPedPdVdQe edRƒedSƒd#d4e dTedTdVdUdKe edŸƒedŸƒd#d/e dVedVdVe ed7ƒed7ƒd#de dWedWdVe ed“ƒed“ƒd#de dXe edXdVedYdVgddZdKe ed+ƒed½ƒed3ƒd”d/e d[e ed[dVed\dVgdd]dKe edëƒedìƒedvƒd”d^e d_ed_dVd`dKe edìƒed{ƒed5ƒd”d^e daedadVdbe edcƒed6ƒd#d4gddededfdgdhgedidje deded e dCddde dDdddgdddge dke edldVedmdVgddndoe ed5ƒed¨ƒd#d•e dpedpdVdqdre ed‘ƒedsƒd#d•e dtedtdVdudre ƒd•e dve e dvdgddwdxe ed²ƒedTƒd#d•e dye dzdd{d|e ed½ƒedHƒd#d•e d}ed}dVd~de ed?ƒed?ƒd#d•e d€e d€ddd‚e edSƒed+ƒd#d•e dƒe edƒdVed„dVgdd…e ed@ƒed@ƒd#d4e d†ed†dVe edSƒedSƒd#de d‡e ed‡dVedˆdVed‰dVedŠdVgdd‹dŒe ed²ƒedrƒd#d•e de dŽddde ed®ƒed²ƒd#d•e d‘e ed‘dVed’dVgdd“d”e edëƒedëƒd#d•e d•ed•dVd–d”e ƒd•e d—e ed—dVed˜dVgdd™dše edŸƒedvƒd#d•e d›e d›ddœde edŸƒedvƒd#d•e džedždVdŸd d¡e edƒed,ƒd#de d¢ed¢dVd£d e ƒd•e d¤ed¤dVd¥d¦e edSƒed5ƒd#d•e d§ed§dVd¨d©e ed@ƒed@ƒd#d•e dªedªdVd«d¬e ed¨ƒed¨ƒd#d•e d­e e d­ded®dVgdd¯e edéƒedéƒd#d4e d°e ed°dVed±dVgdd²d³e edëƒedëƒd#d•e d´e edµdVed¶dVgdd·d¸e ed²ƒedrƒd#d•gdyed¹dºd»d¼gge d½ed½dVe edƒedƒd#d‘e d¾ed¾dVe ed¨ƒed¨ƒd#d‘e d¿ed¿dVe ed-ƒed-ƒd#d‘e dÀedÀdVe ed*ƒed*ƒd#d‘e dÁedÁdVe ed?ƒed?ƒd#d‘e dÂedÂdVe edëƒedëƒd#d‘e dÃedÃdVe ed5ƒed5ƒd#d‘e dÄedÄdVe edKƒedKƒd#d‘e dÅedÅdVe ed,ƒed,ƒd#d‘e dÆedÆdVe edǃedǃd#d‘e dÈedÈdVe ed½ƒed½ƒd#d‘e dÉedÉdVe edʃed˃d#d‘e dÌedÌdVe ed?ƒed?ƒd#d‘e dÍedÍdVe edSƒed΃d#d‘e dÏe edÏdVedÐdVgde edƒedƒd#d‘e dÑedÑdVe ed‘ƒed‘ƒd#d‘e dÒedÒdVe ed‘ƒed‘ƒd#d‘e dÓedÓdVe ed8ƒed8ƒd#d‘e dÔedÔdVe edAƒedƒd#d‘e dÕedÕdVe ed΃ed΃d#d‘e dÖedÖdVe ed׃ed׃d#d‘e dØedØdVe edÙƒedÙƒd#d‘e dÚedÚdVe edÛƒedÛƒd#d‘e dÜedÜdVe ed›ƒed›ƒd#d‘e dÝedÝdVe edÛƒedÛƒd#d‘e dÞedÞdVe ed‘ƒed‘ƒd#d‘e dßedßdVe edàƒedàƒd#d‘e dáedádVe edƒedƒd#d‘e dâedâdVe edãƒed®ƒd#d‘e däedädVe edåƒedåƒd#d‘e dæedædVe edƒedƒd#d‘e dçedçdVe edåƒedåƒd#d‘e dèedèdVe ed*ƒed*ƒd#d‘e déedédVe edÛƒedÛƒd#d‘d„edêdëdìgdí¢e e dîdedïdjgdedêd ed e dCddde dd‚gddde dd„gdÒdde dƒd„gd…dde dDdddgdEdded e dddde dðd‚gddde ddddgdñdded e dCdddgdòddgge d§e ed§dVedódVgde ed®ƒed®ƒd#d‘e d¬ed¬dVe ed½ƒed½ƒd#d‘e dôedôdVdõdöe edTƒed6ƒd#d•‘e d÷e ed÷dVedødVgddùdúe edHƒedÚƒedrƒd”d•‘e d¯e ed¯dVed±dVgde ed²ƒed²ƒd#d‘e d³e ed³dVedûdVgde edíƒedíƒd#d‘e düe dýddþe ed3ƒedrƒd#dZ‘e dÿe ddde ed4ƒed6ƒd#dZ‘e de ddde ed5ƒedVƒd#dZ‘e de ed¸dVeddVgde edHƒed²ƒd#d‘e d·e ed·dVedÁdVed¼dVgdddde ed½ƒedHƒd#d$‘e d¹ed¹dVde edHƒed²ƒd#dï‘e d¾ed¾dVde ed+ƒed²ƒd#dï‘e d e ed dVed¿dVedÅdVedÄdVgdd d de ed3ƒed-ƒd#d$‘e dºe e dºded»dVgdd dde edHƒed²ƒd#d$‘e dÆe edÆdVedÇdVedÈdVe d dgdddde edɃedʃd#d$‘e dÏe edÏdVedÐdVedÑdVgddddde edʃedÒƒd#dN‘e dÔe edÔdVedÖdVedÕdVedËdVedÍdVed×dVgddddêe ed̃edɃd#d$‘e dÙe edÙdVedÎdVgddddêe ed̃edɃd#d$‘e dÚe edÚdVedÛdVgddddqe ed²ƒed½ƒed!ƒd”d$‘e dÜe edÜdVedÝdVgddddqe ed®ƒedSƒedÞƒd”d$‘e dße edßdVedàdVgddddqe edSƒedƒed®ƒd”d$‘e de dddde edIƒed ƒed!ƒd”d•‘e d"e ed"dVedâdVedãdVedædVgdd#d$dêe edäƒedBƒed̃d”d$‘e d%e d%dd&d'dêe ed̃edäƒd(d$‘e dçe edçdVed)dVed*dVed+dVgdd,d-dêe edëƒedìƒedvƒd”d$‘e d.e d.dde ed5ƒedVƒd#dZ‘e d/e dédd0d1e edvƒedëƒd(d•‘e d2e dðde edvƒed5ƒedñƒed̃ed5ƒedíƒdòd‘e d3ed3dVe edBƒedTƒed3ƒd4d‘e d5ed5dVe edvƒedƒd6d‘e d7e dóde edTƒedäƒedíƒedôƒedäƒedõƒdòd‘e d8e ed8dVed9dVgdd:d;e edäƒed̃d#d•‘e ded>dVd?d@e edrƒedAƒd#d•‘e döe edödVed÷dVedBdVedCdVedDdVgddødùdúe edBƒedûƒed̃d”d$‘e dEedEdVdFe ed̃edBƒed̃d”dG‘e dHedHdVdFe edäƒed@ƒed²ƒd”dG‘e düe edüdVedýdVgddþdÿdúe ed@ƒedƒed²ƒd”d$‘e de eddVe dIdgddddúe edƒedƒed®ƒd”d$‘e dJe edJdVedKdVgddúe ed6ƒedÒƒd#dï‘e dLe edLdVedMdVedNdVedOdVedPdVgddQdRdúe edBƒedûƒed̃d”d$‘e dSe edSdVedTdVgddUe ed-ƒed®ƒedVƒedWƒed®ƒedXƒdòdG‘e dYedYdVdUe edvƒed‘ƒedñƒed̃ed5ƒedíƒdòdG‘e dZedZd d[e ƒdZ‘e d e ed dVed dVeddVeddVgdd d\dêe ed6ƒeduƒedʃd”d$‘e de eddVeddVgdd]d^dêe edƒedƒedƒd”d$‘e d_e ed_dVed`dVgdd_dae ed!ƒedbƒd#d•‘e de eddVeddVgdddcee ed̃ed½ƒedôƒd”dŒeee dddedf¡dge ed²ƒed½ƒed!ƒd”d’gd•‘e dhe edhdVedidVgde ed̃edäƒedôƒd”d‘e de eddVeddVedjdVgddkdle edƒedƒedƒd”d•‘e dme edmdVedndVgddodpe edíƒedõƒd#d•‘e de eddVedqdVeddVgddrdse edƒedtƒedƒd”d•‘e due edudVedvdVgde ed²ƒed½ƒed!ƒd”d‘e d¶ed¶dVe edíƒedíƒd#d‘e dwedwdVe edíƒedíƒd#d‘e d5ed5dVdxdydze ed6ƒd«d$‘e d7ed7dVdxdydze ed8ƒd«d$‘e d©e ed¨dVed©dVedªdVgdd{d|d}e edSƒd«d$‘dyed~dd€dgge d‚ed‚dVe ed3ƒed3ƒd#d‘e dƒedƒdVe ed*ƒedVƒd#d‘e d„ed„dVe ƒd‘e d…ed…dVe ed²ƒed!ƒd#d‘e d†ed†dVe edŸƒed®ƒd#d‘e d‡ed‡dVe ed²ƒed!ƒd#d‘e dˆedˆdVe edŸƒed®ƒd#d‘e d‰ed‰dVe ed½ƒedéƒd#d‘e dŠedŠdVd‹e ed-ƒed!ƒd#d4‘e dŒedŒdVde edéƒedŽƒd#d4‘e deddVe ed-ƒed!ƒd#d‘e deddVe edéƒedŽƒd#d‘e d‘ed‘dVe edŸƒedŸƒd#d‘e d’ed’dVd“e edŸƒedŸƒd#d4‘e d”ed”dVe ed*ƒedVƒd#d‘e d•ed•dVe ed–ƒed—ƒd#d‘e d˜ed˜dVe ƒd‘e d™ed™dVe edšƒed›ƒd#d‘e dœedœdVe edHƒedƒd#d‘e džedždVe ed‘ƒed“ƒd#d‘e dŸedŸdVe ed™ƒed›ƒd#d‘e d ed dVe ed*ƒed ƒd#d‘e d¡ed¡dVe ed3ƒed3ƒd#d‘e d¢ed¢dVe edHƒedƒd#d‘e d£e ed£dVed¤dVed¥dVed¦dVed§dVed¨dVgde ed!ƒed"ƒd#d‘e d©e ed©dVedªdVgde ed@ƒedBƒd#d‘e d«e ed«dVed¬dVgde ed6ƒedIƒd#d‘e d­e ed­dVed®dVgde ed3ƒed6ƒd#d‘e d¯e ed¯dVed°dVed±dVed²dVgde ed*ƒed-ƒd#d‘e d³e ed³dVed´dVedµdVed¶dVgde ed3ƒed6ƒd#d‘e d·e ed·dVed¸dVed¹dVgde ed3ƒed6ƒd#d‘e dºedºdVd»e ed*ƒed+ƒd#d4‘e d¼ed¼dVd½e edƒedMƒd#d4‘e d¾ed¾dVe ƒd‘e d¿ed¿dVe ƒd‘e dÀedÀdVdÁe edƒedUƒd#d4‘e dôedôdVdõe edTƒed6ƒd#d4‘e de eddVedÂdVedÃdVedÄdVedÅdVedÆdVgddÇe ed!ƒed"ƒd#d4‘e d>e ed>dVedÈdVgddÉe ed@ƒed5ƒedAƒedBƒd.d4‘e dDe edDdVedÊdVgddËe ed6ƒedGƒedHƒedIƒd.d4‘e dJe edJdVedÌdVgddKe ed3ƒed4ƒed5ƒed6ƒd.d4‘e d'e ed'dVedÍdVedÎdVedÏdVgddÐe ed*ƒed+ƒed,ƒed-ƒd.d4‘e d0e ed0dVedÑdVedÒdVedÓdVgddÔe ed3ƒed4ƒed5ƒed6ƒd.d4‘e d9e ed9dVedÕdVedÖdVgdd×e ed3ƒed4ƒed5ƒed6ƒd.d4‘e d_ed_dVdae ed6ƒedGƒedHƒedIƒd.d4‘e diedidVdke ed3ƒed4ƒed5ƒed6ƒd.d4‘e dØedØdVdÙe ed*ƒedVƒd#d4‘e dÚedÚdVdÛe ed½ƒed*ƒd#d4‘e dmedmdVdÜe ed5ƒed¨ƒd#d4‘e d÷ed÷dVdùe edHƒedÚƒedrƒd”d4‘e d¾ed¾dVe ed¨ƒed¨ƒd#d‘e dÝedÝdVe ed½ƒed½ƒd#d‘e dÞedÞdVe ƒd‘e dßedßdVe ƒd‘e dàedàdVe ed+ƒed²ƒd#d‘e dáedádVe edäƒed̃d#d‘e dâe edâdVedãdVgde ed½ƒedHƒd#d‘e däedädVe edëƒedvƒd#d‘e dåe edådVedædVgde ed3ƒed6ƒd#d‘e dçedçdVe edäƒed̃d#d‘e dèedèdVe ed{ƒedëƒd#d‘e diedidVdje ed+ƒed²ƒd#d4‘e dledldVdme edäƒed̃d#d4‘e doe edodVedédVgddqe ed½ƒedHƒd#d4‘e dpedpdVdêe edëƒedvƒd#d4‘e dse edsdVedëdVgddue ed3ƒed6ƒd#d4‘e dtedtdVdìe edäƒed̃d#d4‘e dxedxdVdíe ed{ƒedëƒd#d4‘e dîedîdVdïe ƒd4‘e dðedðdVdñe ƒd4‘e d†ed†dVdòe edãƒedKƒd#d4‘e d‡ed‡dVdóe ed5ƒed¶ƒd#d4‘e dôedôdVdõe ƒd4‘e döedödVd÷e ƒd4‘e døedødVdùe ed½ƒedúƒd#d4‘e d”ed”dVdûe edƒedüƒd#d4‘e d•ed•dVdýe ed½ƒedüƒd#d4‘e dþedþdVdÿe edƒedSƒd#d4‘e deddVde ƒd4‘e deddVde ƒd4‘e d,ed,dVde edSƒed®ƒd#d4‘e deddVde ƒd4‘e deddVde ed?ƒed?ƒd#d4‘e d ed dVd e ed<ƒed<ƒd#d4‘e d ed dVd e edëƒedëƒd#d4‘e d ed dVde ƒd4‘e deddVde edsƒed<ƒd#d4‘e deddVde ƒd4‘e deddVde edSƒedŸƒd#d4‘e deddVde ƒd4‘e deddVde ƒd4‘e deddVde ƒd4‘e deddVdde edƒed¨ƒd#d/‘e deddVe edãƒedKƒd#d‘e d ed dVe ed‘ƒedƒd#d‘e d!ed!dVe ƒd‘e d"ed"dVd#de edŸƒed*ƒd#d/‘e d$ed$dVe ƒd‘e d¿ed¿dVe ed½ƒed3ƒd#d‘e dÀedÀdVe edSƒed®ƒd#d‘e d%ed%dVe ƒd‘e dÁedÁdVe ed?ƒed?ƒd#d‘e d&ed&dVe ed<ƒed<ƒd#d‘e dÂedÂdVe edëƒedëƒd#d‘e d'ed'dVe ƒd‘e dÃedÃdVe edsƒed<ƒd#d‘e d(ed(dVe ƒd‘e d)ed)dVe ƒd‘e d*ed*dVe ƒd‘e d+ed+dVe ƒd‘e d,ed,dVd-e ed‘ƒed®ƒd#d4‘e dpedpdVdqe ed‘ƒedsƒd#d4‘e dtedtdVdue ƒd4‘e d.ed.dVd/e edŸƒedŸƒd#d4‘e d0ed0dVd1e ed¨ƒed¨ƒd#d4‘e d2ed2dVd3e ƒd4‘e d4ed4dVd5e ƒd4‘e d6ed6dVe ed*ƒedVƒd#d‘e d7ed7dVd8e ed-ƒedrƒd#d4‘e d9ed9dVd:e ed–ƒed—ƒd#d4‘e d;ed;dVded>dVe ed–ƒed—ƒd#d‘e d?ed?dVe ed¨ƒedVƒd#d‘e d@ed@dVe ƒd‘e dAedAdVdBe ed*ƒedVƒd#d4‘e däedädVdCe edSƒedDƒedƒed®ƒd.d4‘e dåedådVdEe ƒd4‘e dãedãdVdFe edìƒed5ƒd#d4‘e dîe edîdVedGdVgddñe ed5ƒedHƒedìƒedäƒd.d4‘e dIedIdVdJe edSƒed®ƒd#d4‘e dþe edþdVedÿdVgddKe edëƒedLƒed{ƒedvƒd.d4‘e deddVdMe edëƒedLƒed{ƒedNƒd.d4‘e deddVdOe edëƒedLƒed{ƒedNƒd.d4‘e de eddVedPdVedQdVedRdVedSdVedTdVgdde edBƒedUƒedVƒed̃d.d4‘e dWedWdVdXe ƒd4‘e d ed dVd e edìƒedYƒedÞƒed5ƒd.d4‘e deddVdZe ed{ƒed[ƒedýƒedëƒd.d4‘e deddVde edBƒedTƒd#d4‘e d\ed\dVd]e ed*ƒed*ƒd#d4‘e d^ed^dVd_e edŸƒedŸƒd#d4‘e d`ed`dVdae ƒd4‘e dbedbdVdce edƒedSƒd#d4‘e ddedddVdee ƒd4‘e d©ed©dVdfe edSƒedŸƒd#d4‘e dgedgdVdhe ƒd4‘e diedidVdje ed8ƒed?ƒd#d4‘e dkedkdVdle ƒd4‘e dmedmdVdne ƒd4‘e doedodVdpe ed“ƒed“ƒd#d4‘e dqedqdVdre ƒd4‘e dsedsdVdte ed“ƒed“ƒd#d4‘e duedudVdve ƒd4‘e dwedwdVdxe ed“ƒed“ƒd#d4‘e dyedydVdze ed̃ed{ƒd#d4‘e d|ed|dVe edSƒed®ƒd#d‘e d}ed}dVe ƒd‘e d~ed~dVe edìƒed5ƒd#d‘e deddVe ƒd‘e d€ed€dVe ƒd‘e de eddVed‚dVgddƒe ed5ƒedäƒd#d4‘e d„ed„dVdûe edSƒed…ƒedƒed®ƒd.d4‘e d†ed†dVd‡e edSƒed®ƒd#d4‘e dˆedˆdVe edëƒedvƒd#d‘e d‰ed‰dVdŠe ed5ƒed‹ƒedìƒedäƒd.d4‘e dŒedŒdVe edëƒedNƒd#d‘e deddVe ƒd‘e dŽedŽdVe edBƒed̃d#d‘e deddVe edìƒed5ƒd#d‘e deddVe ed{ƒedëƒd#d‘e d‘ed‘dVe ƒd‘e d’e ed’dVed“dVgde ed½ƒedHƒd#d‘e d”ed”dVe edBƒedTƒd#d‘e d•ed•dVe ed*ƒed*ƒd#d‘e dÅedÅdVe ed(ƒed(ƒd#d‘e d–ed–dVe ƒd‘e d—ed—dVe edƒedSƒd#d‘e d˜ed˜dVe ƒd‘e d™ed™dVe ƒd‘e dšedšdVe edSƒedŸƒd#d‘e d›ed›dVe ƒd‘e dœedœdVe ed8ƒed?ƒd#d‘e deddVe ƒd‘e džedždVe ƒd‘e dŸe edŸdVed dVgde ed+ƒed²ƒd#d‘e d¡e ed¡dVed¢dVgde ed+ƒed²ƒd#d‘e d·e ed·dVed¼dVgdd£e ed½ƒedHƒd#d4‘e d¹ed¹dVd¤e ed+ƒed²ƒd#d4‘e d¾ed¾dVd¥e ed+ƒed²ƒd#d4‘e d ed dVd e ed3ƒed-ƒd#d4‘e dºedºdVd¦e edHƒed²ƒd#d4‘e dÆe edÆdVedÇdVgdd§e edɃedʃd#d4‘e dÍedÍdVd¨e ed̃edɃd#d4‘e dÔe edÔdVed×dVgdd©e ed̃edɃd#d4‘e dÚedÚdVdªe ed²ƒed½ƒed!ƒd”d4‘e dÜedÜdVd«e ed®ƒedSƒedÞƒd”d4‘e dßedßdVd¬e edSƒedƒed®ƒd”d4‘e deddVde edIƒed ƒed!ƒd”d4‘e d"e ed"dVedãdVedædVgdd­e edäƒedBƒed̃d”d4‘e dâedâdVd®e edTƒed6ƒd#d4‘e dçe edçdVed)dVgdd¯e edëƒedìƒedvƒd”d4‘e d*ed*dVd°e edëƒedvƒd#d4‘e d8ed8dVd:e edäƒed̃d#d4‘e d>ed>dVd?e edrƒedAƒd#d4‘e d±ed±dVd²e ed²ƒed̃d#d4‘e d³ed³dVd´e ed²ƒed̃d#d4‘e dµe edµdVed¶dVgdd·e ed3ƒedMƒed6ƒd”d4‘e d¸e ed¸dVed¹dVgddºe ed5ƒedìƒed½ƒd”d4‘e d»ed»dVd¼e edTƒed6ƒd#d4‘e d½ed½dVd¾e edTƒed6ƒd#d4‘e dÆedÆdVe ed¿ƒed¿ƒd#d‘e dÀedÀdVdÁe edƒed?ƒd#d4‘e dÂedÂdVdÃe edéƒed*ƒd#d4‘e dÄedÄdVdÅe edsƒed5ƒd#d4‘e dÆedÆdVe ƒd‘e dÇedÇdVe ed*ƒedVƒd#d‘e dÈedÈdVdÉe edäƒed̃d#d4‘e dÊedÊdVdËe edäƒed̃d#d4‘e dÌedÌdVe edäƒed̃d#d‘e dÍedÍdVe edäƒed̃d#d‘e dÎedÎdVdÏe ƒd4‘e dÐedÐdVdÑe ƒd4‘e dÒedÒdVdÓe ed²ƒed!ƒd#d4‘e dÔedÔdVdÕe edŸƒed®ƒd#d4‘e dÈedÈdVe ed½ƒed½ƒd#d‘e dÖedÖdVd×e ƒd4‘e dØedØdVdÙe ƒd4‘e dÚedÚdVdÛe edåƒedåƒd#d4‘e dÜedÜdVdÝe ed8ƒedƒd#d4‘e dÞedÞdVdße edéƒed*ƒd#d4‘e dàedàdVdáe edéƒed*ƒd#d4‘e dâedâdVe ed*ƒed¨ƒd#d‘e dãedãdVdäe ed8ƒed8ƒd#d4‘e dåedådVdæe edëƒedëƒd#d4‘e dçedçdVdèe edýƒedýƒd#d4‘e déedédVe ed8ƒed8ƒd#d‘e dêedêdVe edëƒedëƒd#d‘e dëedëdVe edýƒedýƒd#d‘e dìedìdVdíe ed5ƒed®ƒd#d4‘e dîedîdVdïe edAƒed‘ƒd#d4‘e dðedðdVdñe ed-ƒedrƒd#d4‘e dòedòdVdóe edŸƒedBƒd#d4‘e dôedôdVdõe ed²ƒed²ƒd#d4‘e döedödVd÷e ed*ƒed*ƒd#d4‘e døedødVdùe edSƒedƒd#d4‘e dúedúdVd°e edûƒedüƒd#d4‘e dýedýdVdþe ƒd4‘e dÿedÿdVde edäƒed3ƒd#d4‘e deddVde edŸƒedBƒd#d4‘e deddVde edsƒed5ƒd#d4‘e deddVde edvƒedƒd#d4‘e deddVd e ƒd4‘e d ed dVd e ed+ƒed+ƒd#d4‘e d ed dVd e edŸƒedŸƒd#d4‘e deddVde edƒedƒd#d4‘e deddVde ƒd4‘e deddVde edƒedýƒd#d4‘e deddVde ƒd4‘e deddVde edýƒed8ƒd#d4‘e deddVde ƒd4‘e deddVde edVƒedVƒd#d4‘e deddVde edƒed“ƒd#d4‘e deddVd e ƒd4‘e d!ed!dVd"e ƒd4‘e d#ed#dVd$e ed6ƒed®ƒd#d4‘e d%ed%dVd&e ƒd4‘e d'ed'dVd(e edëƒedvƒd#d4‘e d)ed)dVd*e ƒd4‘e d+ed+dVd,e ed,ƒed5ƒd#d4‘e d-ed-dVd.e ƒd4‘e d/ed/dVd0e edŸƒedŸƒd#d4‘e dFedFdVd1e ed8ƒed‘ƒd#d4‘e d2ed2dVdUe edƒedƒd#d4‘e d3ed3dVd4e ed¶ƒed¶ƒd#d4‘e d5ed5dVd6e ed*ƒedVƒd#d4‘e d7ed7dVd8e ed¨ƒed9ƒd#d4‘e d:e ed:dVed;dVgdde ed½ƒedHƒd#d4‘e d?ed?dVd@e ed-ƒedrƒd#d4‘e dAedAdVdBe ed3ƒed3ƒd#d4‘e dCedCdVdDe edäƒed3ƒd#d4‘e d ed dVd e ƒd4‘e dEedEdVe edFƒedFƒd#d‘e dGedGdVdHe ed@ƒed+ƒd#d4‘e dIedIdVe edµƒedµƒd#d‘e dJedJdVe ed¨ƒed¨ƒd#d‘e dÉedÉdVe ed5ƒed®ƒd#d‘e dÌedÌdVe edAƒed‘ƒd#d‘e dKedKdVe ed²ƒed²ƒd#d‘e dLedLdVe ed*ƒed*ƒd#d‘e dMedMdVe ƒd‘e dÍedÍdVe edNƒed΃d#d‘e dÏedÏdVe ed8ƒedAƒd#d‘e dOedOdVe ƒd‘e dÒedÒdVe edƒedƒd#d‘e dPedPdVe ƒd‘e dÓedÓdVe edýƒedýƒd#d‘e dQedQdVe ƒd‘e dÔedÔdVe ed’ƒedƒd#d‘e dRedRdVe ƒd‘e dSedSdVe ed¨ƒed¨ƒd#d‘e dÕedÕdVe edSƒed@ƒd#d‘e dTedTdVe ƒd‘e dUedUdVe edÛƒedUƒd#d‘e dVedVdVe ƒd‘e dWedWdVe ed,ƒed5ƒd#d‘e dXedXdVe ƒd‘e dYedYdVe edŸƒedŸƒd#d‘e dZedZdVe edSƒedSƒd#d‘e d[ed[dVe ed+ƒed+ƒd#d‘e d\ed\dVe edSƒedSƒd#d‘e d]ed]dVe edSƒedSƒd#d‘e d^ed^dVe ed(ƒed›ƒd#d‘e d_ed_dVe edƒedSƒd#d‘e d`ed`dVe ed(ƒed(ƒd#d‘e dÖedÖdVe ed½ƒed½ƒd#d‘e daedadVdbe ed*ƒed*ƒd#d4‘e dcedcdVdde edŸƒedƒd#d4‘e deededVdfe ed5ƒedgƒd#d4‘e dhedhdVdie edƒedjƒd#d4‘e dkedkdVe edŸƒedƒd#d‘e d}ed}dVd~e ed?ƒed?ƒd#d4‘e d€ed€dVdle edSƒedSƒd#d4‘e dme edmdVedndVgddoe edpƒedàƒd#d4‘e dqedqdVdre edRƒedMƒd#d4‘e dsedsdVdte edŸƒedŸƒd#d4‘e duedudVdve ƒd4‘e d‡e ed‡dVed‰dVedŠdVgdd‹e ed²ƒedrƒd#d4‘e dŽedŽdVdwe edxƒedyƒd#d4‘e deddVde ed®ƒed²ƒd#d4‘e d‘ed‘dVdze edýƒed{ƒd#d4‘e d•ed•dVd–e ƒd4‘e d—ed—dVd|e edŸƒedvƒd#d4‘e d›ed›dVdœe edŸƒedvƒd#d4‘e džedždVdŸe edƒedǃd#d4‘e d¢ed¢dVd£e ƒd4‘e d}ed}dVd~e edƒedëƒd#d4‘e deddVd€e ƒd4‘e deddVd‚e ƒd4‘e d§ed§dVd¨e ed@ƒed@ƒd#d4‘e dƒedƒdVe edSƒedSƒd#d‘e d„ed„dVe ed5ƒed¨ƒd#d‘e d…ed…dVe ed@ƒed@ƒd#d‘e d†ed†dVe ed?ƒed?ƒd#d‘e d‡ed‡dVe edSƒedSƒd#d‘e dØe edØdVedˆdVgde edüƒedÙƒd#d‘e d‰ed‰dVe edŸƒedŸƒd#d‘e dŠedŠdVe edŸƒedŸƒd#d‘e d‹ed‹dVe ƒd‘e dŒe edŒdVeddVedŽdVgde ed²ƒedrƒd#d‘e deddVe edxƒedyƒd#d‘e dÚedÚdVe ed™ƒed,ƒd#d‘e deddVe ƒd‘e d‘ed‘dVe edŸƒedvƒd#d‘e d’ed’dVe edŸƒedvƒd#d‘e d“ed“dVe ed(ƒed›ƒd#d‘e d”ed”dVe ƒd‘e d•ed•dVe edSƒed5ƒd#d‘e d–ed–dVe ƒd‘e d—ed—dVe ed@ƒed@ƒd#d‘e d˜ed˜dVe ed¨ƒed¨ƒd#d‘e d™ed™dVe ed“ƒed“ƒd#d‘e dšedšdVe edSƒedSƒd#d‘e d›ed›dVe ed²ƒedrƒd#d‘e dªedªdVd«e ed¨ƒed¨ƒd#d4‘e dœedœdVde ed,ƒed“ƒd#d4‘e džedždVdŸe ed*ƒedVƒd#d4‘e d ed dVd¡e ed*ƒedVƒd#d4‘e d¢ed¢dVd£e ed*ƒedVƒd#d4‘e d¤ed¤dVe ƒd‘e d¥ed¥dVe ƒd‘e d¦ed¦dVd§e ed*ƒedVƒd#d4‘e d¨ed¨dVd©e ed¿ƒed¿ƒd#d4‘e dªedªdVe ed«ƒed¬ƒd#d‘e d­e ed­dVed®dVgde ed¯ƒedMƒd#d‘e d°ed°dVe edHƒedƒd#d‘e d±ed±dVe ed¯ƒedMƒd#d‘e d²ed²dVe edMƒedHƒd#d‘e d³ed³dVd´e edBƒed²ƒd#d4‘e dµedµdVe ed*ƒedVƒd#d‘e d¶ed¶dVd·e edvƒedvƒd#d4‘e d¸ed¸dVe ƒd‘e dÜedÜdVe edƒed?ƒd#d‘e d¹ed¹dVe ed*ƒed*ƒd#d‘e dºedºdVe edsƒed5ƒd#d‘e d»ed»dVe edvƒedvƒd#d‘e dÝedÝdVe ed<ƒed<ƒd#d‘e dŽedŽdVde ed‘ƒed“ƒd#d4‘e d–ed–dVd—e ed™ƒed›ƒd#d4‘e dœedœdVde ed*ƒed ƒd#d4‘e d¼ed¼dVe edsƒed5ƒd#d‘e d½ed½dVe ƒd‘e d¾ed¾dVe ƒd‘e d¿ed¿dVe ƒd‘e dÀedÀdVe ƒd‘e d e ed dVeddVeddVgdd e ed6ƒeduƒedʃd”d4‘e de eddVeddVgdde edƒedƒedƒd”d4‘e d_ed_dVd_e ed!ƒedbƒd#d4‘e deddVde ed²ƒed½ƒed!ƒd”d4‘e de eddVedjdVgddke edƒedƒedƒd”d4‘e dmedmdVdoe edíƒedõƒd#d4‘e de eddVedqdVgddÁe edƒedtƒedƒd”d4‘e dÂedÂdVe ƒd‘e dÃedÃdVe ƒd‘e dÄedÄdVe edTƒed6ƒd#d‘e dÅe edÅdVedÆdVgde ed½ƒedHƒd#d‘e dÇedÇdVe ed+ƒed²ƒd#d‘e dÈedÈdVe ed+ƒed²ƒd#d‘e dÉedÉdVe ed3ƒed-ƒd#d‘e dÊedÊdVe edHƒed²ƒd#d‘e dËe edËdVedÌdVgde edɃedʃd#d‘e dÍedÍdVe ed̃edɃd#d‘e dÎe edÎdVedÏdVgde edʃedÒƒd#d‘e dÐe edÐdVedÑdVgde ed̃edɃd#d‘e dÒedÒdVe ed²ƒed!ƒd#d‘e dÓedÓdVe ed®ƒedÞƒd#d‘e dÔedÔdVe edSƒed®ƒd#d‘e dÕedÕdVe edIƒed!ƒd#d‘e dÖe edÖdVed×dVedØdVedÙdVgde edäƒed̃d#d‘e dÚedÚdVe edTƒed6ƒd#d‘e dÛe edÛdVedÜdVedÝdVgde edëƒedvƒd#d‘e dÞedÞdVe edrƒedAƒd#d‘e dße edßdVedàdVedádVgde ed6ƒedʃd#d‘e dâe edâdVedãdVgde edƒedƒd#d‘e däedädVe ed!ƒedbƒd#d‘e dåedådVe ed̃edôƒd#d‘e dæe edædVedçdVgde edƒedƒd#d‘e dèe edèdVedédVgde edƒedƒd#d‘e dÞedÞdVe ed(ƒed(ƒd#d‘e dêedêdVdëe ed–ƒed—ƒd#d4‘e dìedìdVe ed+ƒed+ƒd#d‘e díedídVe edŸƒedŸƒd#d‘e dîedîdVe ed²ƒed!ƒd#d‘e dïedïdVe ed+ƒed+ƒd#d‘e dðedðdVe ed²ƒed!ƒd#d‘e dñedñdVe ed3ƒed6ƒd#d‘e dòedòdVe ed+ƒedTƒd#d‘e dóedódVe ed²ƒed!ƒd#d‘e deddVde ed+ƒed+ƒd#d4‘e deddVd e edSƒedSƒd#d4‘e d"ed"dVd#e edSƒedSƒd#d4‘e d%ed%dVd&e ed(ƒed›ƒd#d4‘e d)ed)dVd*e edƒedSƒd#d4‘e d,ed,dVd-e ed(ƒed/ƒd#d4‘e d0ed0dVd1e ƒd4‘e d2ed2dVd3e ƒd4‘e d°ed°dVd²e edSƒedSƒd#d4‘e d¶ed¶dVd·e ed²ƒedrƒd#d4‘e dôedôdVe edõƒedFƒd#d‘e döedödVd÷e ed¨ƒed¨ƒd#d4‘e døedødVdùe edsƒed΃d#d4‘e dúedúdVdûe ƒd4‘e düedüdVdýe ed?ƒedSƒd#d4‘e dþedþdVdÿe ed‘ƒedëƒd#d4‘e deddVde ƒd4‘e deddVde edŸƒedŸƒd#d4‘e deddVde edÞƒedƒedƒd”d4‘e deddVd e ed®ƒed ƒedVƒd”d4‘e d ed dVd e edƒed8ƒedŸƒd”d4‘e d ed dVde ed*ƒed ƒd#d4‘e deddVde edƒedƒd#d4‘e deddVe ed¨ƒed¨ƒd#d‘e dâedâdVe edsƒed΃d#d‘e deddVe ƒd‘e deddVe edƒedSƒd#d‘e deddVe ƒd‘e deddVe ed(ƒedëƒd#d‘e deddVe ƒd‘e deddVe edvƒedvƒd#d‘e deddVe edŸƒedŸƒd#d‘e deddVe ƒd‘e deddVe edÞƒedƒd#d‘e deddVe ed®ƒedVƒd#d‘e deddVe edƒedŸƒd#d‘e deddVe ed*ƒed ƒd#d‘e d ed dVe edƒedƒd#d‘e d!ed!dVe edŸƒedŸƒd#d‘e d"ed"dVe ed¨ƒed¨ƒd#d‘e d#ed#dVe ƒd‘e d$ed$dVe ƒd‘e d%ed%dVe edéƒedéƒd#d‘e d&ed&dVe ƒd‘e d'ed'dVe edëƒedŸƒd#d‘e d(ed(dVe edŸƒedŸƒd#d‘e d)ed)dVe ƒd‘e d*ed*dVe ƒd‘e d+ed+dVd,e ed¨ƒed¨ƒd#d4‘e d-ed-dVd.e ƒd4‘e d/ed/dVd0e ed@ƒedMƒd#d4‘e d1ed1dVd2e ƒd4‘e d3ed3dVd4e ed‘ƒed“ƒd#d4‘e d5ed5dVd6e ƒd4‘e d»ed»dVd7e ed6ƒedvƒd#d4‘e d8ed8dVd9e ƒd4‘e d:ed:dVd;e ed,ƒedcƒd#d4‘e ded>dVd?e edSƒed5ƒd#d4‘e d@ed@dVdAe ƒd4‘e dBedBdVdCe ed™ƒedDƒd#d4‘e dEedEdVdFe ƒd4‘e dGedGdVdHe ƒd4‘e dIedIdVdJe edëƒedŸƒd#d4‘e dKedKdVdLe ƒd4‘e dMedMdVdNe edŸƒedŸƒd#d4‘e dOedOdVdPe ƒd4‘e dQedQdVdRe ed²ƒed!ƒd#d4‘e dSedSdVe ed¬ƒed¬ƒd#d‘e dTedTdVdUe ƒd4‘e dVedVdVe ƒd‘e dWedWdVdXe ed*ƒedVƒd#d4‘e dYedYdVdZe ed@ƒed½ƒd#d4‘e d[ed[dVe ed*ƒedVƒd#d‘e d\ed\dVe edåƒedåƒd#d‘e d]ed]dVe ed8ƒedƒd#d‘e d^ed^dVe edéƒed*ƒd#d‘e d_ed_dVe edéƒed*ƒd#d‘e d`ed`dVdae ƒd4‘e dbedbdVe ed¶ƒed¶ƒd#d‘e dcedcdVe ed<ƒed<ƒd#d‘e ddedddVe ƒd‘e deededVdfe ƒd4‘e dgedgdVdhe ed½ƒed*ƒd#d4‘e diedidVdje ed+ƒed+ƒd#d4‘e dkedkdVdle ed²ƒed!ƒd#d4‘e dmedmdVdne ed3ƒed6ƒd#d4‘e doedodVdpe ed+ƒedTƒd#d4‘e dqedqdVdre ed²ƒed!ƒd#d4‘e dèedèdVe ed*ƒed*ƒd#d‘e dsedsdVe ƒd‘e dtedtdVdue ed¬ƒed¬ƒd#d4‘e dvedvdVdwe ed<ƒed<ƒd#d4‘e dxedxdVe edéƒedyƒd#d‘e dzedzdVe ed*ƒed+ƒd#d‘e d{ed{dVe ed@ƒed½ƒd#d‘e d|ed|dVe ed½ƒed*ƒd#d‘e d}ed}dVe ed½ƒed½ƒd#d‘e d~ed~dVe ƒd‘e deddVe ƒd‘e d€ed€dVde ed*ƒedVƒd#d4‘e d‚ed‚dVe edõƒedFƒd#d‘e dƒedƒdVe ed(ƒed(ƒd#d‘e d„ed„dVd…e edƒedƒd#d4‘e d†ed†dVd‡e ed½ƒed*ƒd#d4‘e dˆedˆdVd‰e edMƒedVƒd#d4‘e dŠedŠdVd‹e ed½ƒed*ƒd#d4‘e dŒedŒdVde edHƒedHƒd#d4‘e d9ed9dVd:e ed<ƒed<ƒd#d4‘e dŽedŽdVe ed²ƒed̃d#d‘e deddVe ed²ƒed̃d#d‘e deddVe ed3ƒed6ƒd#d‘e d‘ed‘dVe ed5ƒed½ƒd#d‘e d’ed’dVe edTƒed6ƒd#d‘e d“ed“dVe edTƒed6ƒd#d‘e d”ed”dVe ed4ƒed4ƒd#d‘e d•ed•dVd–e ed3ƒed3ƒd#d4‘e d—ed—dVdKe ed/ƒed˜ƒd#dï‘e d™ed™dVdše ed®ƒed›ƒd#dï‘d„edœddždŸgd e d ed dVd d¡e ed+ƒed+ƒd#d•e d ed dVd d¡e edŸƒedŸƒd#d•e dQedQdVdRd¢e ed²ƒed!ƒd#d•e diedidVdjd£e ed+ƒed+ƒedƒd¤d•e dkedkdVdld¥e ed²ƒed!ƒd#d•e dmedmdVdnd¦e ed3ƒed6ƒed§ƒd¤d•e doedodVdpd¨e ed+ƒedTƒedƒd¤d•e dqedqdVdrd©e ed²ƒed!ƒed§ƒd¤d•gd=edªd«d¬d­ge ed®dVed¯dVgdge d°ed°dVe ed5ƒed5ƒd#d‘e d±ed±dVe ed5ƒed5ƒd#d‘e d²ed²dVe ed5ƒed5ƒd#d‘e d³ed³dVe edŸƒedŸƒd#d‘e d´ed´dVe ed¨ƒed¨ƒd#d‘e dµedµdVe ed¨ƒed¨ƒd#d‘e d¶ed¶dVe ed*ƒed*ƒd#d‘e d·ed·dVe ed¸ƒed¸ƒd#d‘e d¹ed¹dVe edŸƒedŸƒd#d‘e dºedºdVe edŸƒedŸƒd#d‘e d»e ed»dVed¼dVgde edSƒedSƒd#d‘e d½e ed½dVed¾dVgde edSƒedSƒd#d‘e d¿e ed¿dVedÀdVgde ed5ƒed5ƒd#d‘e dÁe edÁdVedÂdVgde edSƒedSƒd#d‘e dÃedÃdVe ed¨ƒed¨ƒd#d‘e dÄe edÄdVedÅdVgde edŸƒedŸƒd#d‘e dÆedÆdVe ed5ƒed5ƒd#d‘e dÇedÇdVe edŸƒedŸƒd#d‘e dÈedÈdVe ed5ƒed5ƒd#d‘e dÉe edÉdVedÊdVedËdVgde edŸƒedŸƒd#d‘e dÌedÌdVe ed¸ƒed¸ƒd#d‘e dÍedÍdVe ed΃ed΃d#d‘e dÏedÏdVe ed¨ƒed¨ƒd#d‘e dÐedÐdVe edŸƒedŸƒd#d‘e dÑedÑdVe ed*ƒed*ƒd#d‘e dÒedÒdVe ed¨ƒed¨ƒd#d‘e dÓe edÓdVedÔdVgde edSƒedSƒd#d‘e dÕe edÕdVedÖdVgde edŸƒedŸƒd#d‘e d×ed×dVe ed5ƒed5ƒd#d‘e dØedØdVe edŸƒedŸƒd#d‘e dÙedÙdVe ed¸ƒed¸ƒd#d‘e dÚedÚdVe ed¨ƒed¨ƒd#d‘e dÛedÛdVe edŸƒedŸƒd#d‘e dÜedÜdVe ed¨ƒed¨ƒd#d‘e dÝedÝdVe edŸƒedŸƒd#d‘e dÞedÞdVe ed¶ƒed¶ƒd#d‘e dßedßdVe ed‘ƒedUƒd#d‘e dàedàdVe ed<ƒed6ƒd#d‘e dáedádVe edâƒedâƒd#d‘e dãedãdVe ed¶ƒed¶ƒd#d‘e däedädVe edSƒedSƒd#d‘e dåedådVe ed<ƒed<ƒd#d‘e dæedædVe edNƒedNƒd#d‘e dçedçdVe ed¶ƒed¶ƒd#d‘e dèedèdVe ed<ƒed<ƒd#d‘e déedédVe ed¶ƒed¶ƒd#d‘e dêedêdVe edVƒedVƒd#d‘e dëedëdVe edSƒedSƒd#d‘e dìedìdVe edŸƒedŸƒd#d‘e díedídVe edŸƒedŸƒd#d‘e dîedîdVe edŸƒedŸƒd#d‘e dïedïdVe edñƒedñƒd#d‘e dðedðdVe ed¨ƒed¨ƒd#d‘e dñedñdVe ed¨ƒed¨ƒd#d‘e dòedòdVe edŸƒedŸƒd#d‘e dóedódVe edŸƒedŸƒd#d‘e dôedôdVe edŸƒedŸƒd#d‘e dõedõdVe edŸƒedŸƒd#d‘e döedödVe edŸƒedŸƒd#d‘e d÷ed÷dVe edŸƒedŸƒd#d‘e døedødVe edŸƒedŸƒd#d‘e dùedùdVe edŸƒedŸƒd#d‘e dúedúdVe edŸƒedŸƒd#d‘e dûedûdVe edSƒedSƒd#d‘e düedüdVe edSƒedSƒd#d‘e dýedýdVe edSƒedSƒd#d‘e dþedþdVe edŸƒedŸƒd#d‘e dÿedÿdVe edŸƒedŸƒd#d‘e deddVe edŸƒedŸƒd#d‘e deddVe ed5ƒed5ƒd#d‘e deddVe ed*ƒed*ƒd#d‘e deddVe edŸƒedŸƒd#d‘deddddged d ed dVed e dCddde dd‚gddde dƒd„gd…dde dDdddgdEddge d±e ed±dVed dVed dVgdd²d d\e ed²ƒed̃d#d$e d³e ed³dVeddVeddVgdd´dd\e ed²ƒed̃d#d$e dµe edµdVeddVed¶dVgdd·ddKe ed3ƒedMƒed6ƒd”d$e de eddVeddVeddVgddddKe edTƒedBƒedƒd”d$e d¸e ed¸dVed¹dVeddVgddºddKe ed5ƒedìƒed½ƒd”d$e de eddVeddVeddVgddddKe edvƒedëƒed-ƒd”d$e de eddVed dVed!dVgdd"d#d$e ed3ƒedMƒed6ƒd”d$gdygZeeed%<dS(&z?DO NOT EDIT THIS FILE DIRECTLY, INSTEAD RUN `make package-data`éN)ÚDecimalé)Ú*)Ú providersÚ anthropicÚ Anthropiczhttps://api\.anthropic\.comz%https://www.anthropic.com/pricing#apiÚclaude)ÚcontainsÚusageÚ input_tokensT)ÚpathÚdestÚrequiredÚcache_creation_input_tokensFÚcache_read_input_tokensÚcache_write_tokensÚcache_read_tokensÚ output_tokensÚdefaultÚmodel)ÚrootÚmappingsÚ api_flavorÚ model_pathzclaude-2)Ú starts_withz claude-v2)Úor_zClaude 2.0 / 2.1zClaude 2 is Anthropic's previous generation model, offering reliable performance for various tasks. This includes Claude 2.0 and Claude 2.1. i@ Ú8Ú24)Ú input_mtokÚ output_mtok)ÚidÚmatchÚnameÚ descriptionÚcontext_windowÚpriceszclaude-3-5-haiku-latestzclaude-3-5-haikuzclaude-3.5-haikuzClaude Haiku 3.5z"Fastest, most cost-effective modelz0.8Ú1z0.08Ú4)rÚcache_write_mtokÚcache_read_mtokrzclaude-3-5-sonnetzclaude-3.5-sonnetzClaude Sonnet 3.5z¨Claude 3.5 Sonnet is an ideal balance of intelligence and speed for enterprise workloads. Maximum utility at a lower price, dependable, balanced for scaled deployments.Ú3z3.75z0.3Ú15zclaude-3-7-sonnet-latestzclaude-3-7-sonnetzclaude-3.7-sonnetzclaude-sonnet-3.7zclaude-sonnet-3-7zClaude Sonnet 3.7zxClaude 3.7 Sonnet is an advanced large language model with improved reasoning, coding, and problem-solving capabilities.zclaude-3-haikuzClaude Haiku 3z0.25z0.03z1.25zclaude-3-opus-latestz claude-3-opusz Claude Opus 3z–Claude 3 Opus was Anthropic's most powerful model for highly complex tasks. It boasts top-level performance, intelligence, fluency, and understanding.z18.75z1.5Ú75zclaude-3-sonnetzClaude 3 Sonnetz¦Claude 3 Sonnet is an ideal balance of intelligence and speed for enterprise workloads. Maximum utility at a lower price, dependable, balanced for scaled deployments.zclaude-haiku-4-5zclaude-haiku-4.5zclaude-4-5-haikuzclaude-4.5-haikuzClaude Haiku 4.5z(Fastest and most intelligent Haiku modelz0.1Ú5zclaude-instant-1)Úequalsz$Retired, here to match price sourcesz1.63z55.1)r r!r#r%zclaude-instant-1.2z5.51zclaude-opus-4-0z claude-4-opusz claude-opus-4zclaude-opus-4-20250514z Claude Opus 4z(Most intelligent model for complex taskszclaude-opus-4-1zclaude-opus-4.1zClaude Opus 4.1zclaude-sonnet-4-0zclaude-sonnet-4-2025zclaude-sonnet-4@zclaude-sonnet-4zclaude-4-sonnetzClaude Sonnet 4z0Optimal balance of intelligence, cost, and speedzclaude-sonnet-4-5zclaude-sonnet-4.5zClaude Sonnet 4.5z5Most intelligent model for building agents and codingi@BÚ6)ÚstartÚprice)ÚbaseÚtiersz7.5z0.6z22.5z claude-v1)r r"Ú api_patternÚ pricing_urlsÚ model_matchÚprovider_matchÚ extractorsÚmodelsÚavianÚAvianzhttps://api\.avian\.iozhttps://avian.io/pricing/zMeta-Llama-3.1-405B-Instruct)r r!r%zMeta-Llama-3.1-70B-Instructz0.45zMeta-Llama-3.1-8B-InstructzMeta-Llama-3.3-70B-Instruct)r r"r4r5r9Úawsz AWS Bedrockz4https://bedrock-runtime\.[a-z0-9-]+\.amazonaws\.com/z'https://aws.amazon.com/bedrock/pricing/ÚbedrockÚ inputTokensÚ outputTokenszmeta.llama3-8b-instruct-v1:0zllama3-8b-instruct-v1z nova-lite-v1z Nova Lite 1.0a&Amazon Nova Lite 1.0 is a very low-cost multimodal model from Amazon that focused on fast processing of image, video, and text inputs to generate text output. Amazon Nova Lite can handle real-time customer interactions, document analysis, and visual question-answering tasks with high accuracy.z0.06z0.015z0.24)rr)r)r r!r"r#r%z nova-micro-v1zNova Micro 1.0a˜Amazon Nova Micro 1.0 is a text-only model that delivers the lowest latency responses in the Amazon Nova family of models at a very low cost. With a context length of 128K tokens and optimized for speed and cost, Amazon Nova Micro excels at tasks such as text summarization, translation, content classification, interactive chat, and brainstorming. It has simple mathematical reasoning and coding abilities.z0.035z0.00875z0.14z nova-pro-v1z Nova Pro 1.0a7Amazon Nova Pro 1.0 is a capable multimodal model from Amazon focused on providing a combination of accuracy, speed, and cost for a wide range of tasks. As of December 2024, it achieves state-of-the-art performance on key benchmarks including visual question answering (TextVQA) and video understanding (VATEX).z0.2z3.2)r r"r4r5r7r8r9ÚazurezMicrosoft AzurezY(https?://)?([^.]*\.)?(?:openai\.azure\.com|azure-api\.net|cognitiveservices\.azure\.com)z\https://azure.microsoft.com/en-us/pricing/details/cognitive-services/openai-service/#pricingzÍThese are prices for "*-Global" models, prices for "Regional" models are often slightly higher. Retired models are listed at https://learn.microsoft.com/th-th/azure/ai-foundry/openai/concepts/legacy-modelsÚadaztext-embedding-adaztext-embedding-ada-002ztext-embedding-ada-002-v2)rÚbabbagez babbage-002z0.4Úcuriez text-curieztext-curie-001Ú2Údavinciz davinci-002z text-davinciztext-davinci-002z gpt-3.5-turbozgpt-3.5-turbo-0301zgpt-3.5-turbo-0613zgpt-3.5-turbo-instructzgpt-3.5-turbo-instruct-0914zgpt-3.5-turbo-0125z0.5zgpt-3.5-turbo-1106zgpt-3.5-turbo-16k-0613z gpt-35-16kz gpt-35-turbozgpt-35-turbo-0613zgpt-35-turbo-1106zgpt-35-turbo-16kzgpt-35-turbo-16k-0613zgpt-4z gpt-4-0314z gpt-4-0613Ú30Ú60zgpt-4-0125-previewÚ10zgpt-4-1106-previewzgpt-4-1106-vision-previewz gpt-4-32kzgpt-4-32k-0314zgpt-4-32k-0613Ú120zgpt-4-preview-1106z gpt-4-turbozgpt-4-turbo-0125-previewzgpt-4-turbo-2024-04-09zgpt-4-turbo-previewz gpt-4-visionzgpt-4-vision-previewzgpt-4.1zgpt-4.1-2025-04-14z gpt-4.1-minizgpt-4.1-mini-2025-04-14z1.6z gpt-4.1-nanozgpt-4.1-nano-2025-04-14zgpt-4o-2024-0513zgpt-4o-2024-05-13zgpt-4o-2024-08-06z2.5zgpt-4o-2024-1120zgpt-4o-2024-11-20z gpt-4o-minizgpt-4o-mini-0718zgpt-4o-mini-audioiôz0.15z0.075Ú20)rr)rÚinput_audio_mtokÚoutput_audio_mtok)r r!r$r%zgpt-4o-mini-realtimez2.4)rr)rrKÚcache_audio_read_mtokrLzgpt-4o-realtimeÚ40Ú80zgpt-5zgpt-5-2025-08-07zGPT-5zYGPT-5 is OpenAI's flagship model for coding, reasoning, and agentic tasks across domains.i€z0.125z gpt-5-minizgpt-5-mini-2025-08-07z GPT-5 miniztGPT-5 mini is a faster, more cost-efficient version of GPT-5. It's great for well-defined tasks and precise prompts.z0.025z gpt-5-nanozgpt-5-nano-2025-04-14z GPT-5 nanozqGPT-5 Nano is OpenAI's fastest, cheapest version of GPT-5. It's great for summarization and classification tasks.z0.05z0.005Úgpt35zgpt4-turbo-previewzmai-ds-r1:freezMAI DS R1 (free)aÿMAI-DS-R1 is a post-trained variant of DeepSeek-R1 developed by the Microsoft AI team to improve the model's responsiveness on previously blocked topics while enhancing its safety profile. Built on top of DeepSeek-R1's reasoning foundation, it integrates 110k examples from the Tulu-3 SFT dataset and 350k internally curated multilingual safety-alignment samples. The model retains strong reasoning, coding, and problem-solving capabilities, while unblocking a wide range of prompts previously restricted in R1.Úo1z o1-2024-12-17z o1-previewzo1-preview-2024-09-12zo1-minizo1-mini-2024-09-12z1.1z0.55z4.4z o3-2025-04-16Úo3zo3-minizo3-mini-2025-01-31zo4-minizo4-mini-2025-04-16z0.28zphi-3-medium-128k-instructzPhi-3 Medium 128K Instructa3Phi-3 128K Medium is a powerful 14-billion parameter model designed for advanced language understanding, reasoning, and instruction following. Optimized through supervised fine-tuning and preference adjustments, it excels in tasks involving common sense, mathematics, logical reasoning, and code processing.zphi-3-mini-128k-instructzPhi-3 Mini 128K Instructa&Phi-3 Mini is a powerful 3.8B parameter model designed for advanced language understanding, reasoning, and instruction following. Optimized through supervised fine-tuning and preference adjustments, it excels in tasks involving common sense, mathematics, logical reasoning, and code processing.zphi-3.5-mini-128k-instructzPhi-3.5 Mini 128K InstructavPhi-3.5 models are lightweight, state-of-the-art open models. These models were trained with Phi-3 datasets that include both synthetic data and the filtered, publicly available websites data, with a focus on high quality and reasoning-dense properties. Phi-3.5 Mini uses 3.8B parameters, and is a dense decoder-only transformer model using the same tokenizer as Phi-3 Mini.zphi-4zPhi 4z²Microsoft Research Phi-4 is designed to perform well in complex reasoning tasks and can operate efficiently in situations with limited memory or where quick responses are needed.z0.07zphi-4-multimodal-instructzPhi 4 Multimodal Instructa&Phi-4 Multimodal Instruct is a versatile 5.6B parameter foundation model that combines advanced reasoning and instruction-following capabilities across both text and visual inputs, providing accurate text outputs. The unified architecture enables efficient, low-latency inference, suitable for edge and mobile deployments. Phi-4 Multimodal Instruct supports text inputs in multiple languages including Arabic, Chinese, English, French, German, Japanese, Spanish, and more, with visual input optimized primarily for English. It delivers impressive performance on multimodal tasks involving mathematical, scientific, and document reasoning, providing developers and enterprises a powerful yet compact model for sophisticated interactive applications. For more information, see the Phi-4 Multimodal blog post.zphi-4-reasoning-pluszPhi 4 Reasoning Plusa{Phi-4-reasoning-plus is an enhanced 14B parameter model from Microsoft, fine-tuned from Phi-4 with additional reinforcement learning to boost accuracy on math, science, and code reasoning tasks. It uses the same dense decoder-only transformer architecture as Phi-4, but generates longer, more comprehensive outputs structured into a step-by-step reasoning trace and final answer.z0.35zphi-4-reasoning-plus:freezPhi 4 Reasoning Plus (free)zphi-4-reasoning:freezPhi 4 Reasoning (free)aÚPhi-4-reasoning is a 14B parameter dense decoder-only transformer developed by Microsoft, fine-tuned from Phi-4 to enhance complex reasoning capabilities. It uses a combination of supervised fine-tuning on chain-of-thought traces and reinforcement learning, targeting math, science, and code reasoning tasks. With a 32k context window and high inference efficiency, it is optimized for structured responses in a two-part format: reasoning trace followed by a final solution.ztext-embedding-3-largez0.13ztext-embedding-3-smallz0.02zwizardlm-2-8x22bzWizardLM-2 8x22BzæWizardLM-2 8x22B is Microsoft AI's most advanced Wizard model. It demonstrates highly competitive performance compared to leading proprietary models, and it consistently outperforms all existing state-of-the-art opensource models.z0.48)r r"r4r5Úprice_commentsr9ÚcerebrasÚCerebraszhttps://api\.cerebras\.aiz'https://www.cerebras.ai/pricing#pricingz4https://inference-docs.cerebras.ai/models/openai-ossÚ prompt_tokensÚcompletion_tokensÚchatz gpt-oss-120bzcerebras/gpt-oss-120bzcerebras:gpt-oss-120bz GPT-OSS 120BzþOpenAI's flagship open source model, built on a Mixture-of-Experts (MoE) architecture with 120 billion parameters and 128 experts. Delivers frontier reasoning capabilities with record-breaking inference speeds on Cerebras hardware (~3,000 tokens/second).izHDeveloper tier pricing. Free tier: 65k context, Paid tier: 131k context.z0.75)r r!r"r#r$rSr%z llama-3.3-70bzcerebras/llama-3.3-70bzcerebras:llama-3.3-70bz Llama 3.3 70BzÎMeta's enhanced 70B model delivering 405B-level accuracy. Optimized for chat, coding, instruction following, mathematics, and reasoning with high-speed inference on Cerebras hardware (~2,100 tokens/second).zHDeveloper tier pricing. Free tier: 65k context, Paid tier: 128k context.z0.85z1.2z llama3.1-8bzcerebras/llama3.1-8bzcerebras:llama3.1-8bz Llama 3.1 8Bz°Meta's Llama 3.1 8B model for general-purpose tasks including chat, coding, and instruction following. Optimized for fast inference on Cerebras hardware (~2,200 tokens/second).i€zFDeveloper tier pricing. Free tier: 8k context, Paid tier: 32k context.z qwen-3-32bzcerebras/qwen-3-32bzcerebras:qwen-3-32bz Qwen 3 32BzÑQwen's 32B parameter model with enhanced reasoning and coding capabilities. Supports both standard and reasoning modes for complex tasks, with fast inference speeds on Cerebras hardware (~2,600 tokens/second).ÚcohereÚCoherezhttps://api\.cohere\.aizhttps://cohere.com/pricingzcommand-Ú billed_unitsÚcommandÚCommandz±Command is an instruction-following conversational model that performs language tasks with high quality, more reliably and with a longer context than our base generative models.z command-az Command AaiCommand A is an open-weights 111B parameter model with a 256k context window focused on delivering great performance across agentic, multilingual, and coding use cases. Compared to other leading proprietary and open-weights models Command A delivers maximum performance with minimum hardware costs, excelling on business-critical agentic and multilingual tasks.z command-rzcommand-r-08-2024z Command RaCommand-R is a 35B parameter model that performs conversational language tasks at a higher quality, more reliably, and with a longer context than previous models. It can be used for complex workflows like code generation, retrieval augmented generation (RAG), tool use, and agents.zcommand-r-pluszcommand-r-plus-08-2024z Command R+z“Command R+ is a new, 104B-parameter LLM from Cohere. It's useful for roleplay, general consumer usecases, and Retrieval Augmented Generation (RAG).z command-r7bzcommand-r7b-12-2024z Command R7BzÈCommand R7B (12-2024) is a small, fast update of the Command R+ model, delivered in December 2024. It excels at RAG, tool use, agents, and similar tasks requiring complex reasoning and multiple steps.z0.0375ÚdeepseekÚDeepseekzhttps://api\.deepseek\.comz1https://api-docs.deepseek.com/quick_start/pricingzæDeepseek off-peak pricing applies "UTC 16:30-00:30" so we switch it around and use the off-peak pricing as the default (first) price then the second price with a constraint is the "standard" pricing that applies "UTC 00:30-16:30".Úprompt_tokens_detailsÚ cached_tokensÚcompletion_tokens_detailsÚ audio_tokensÚoutput_audio_tokensz deepseek-chatzdeepseek-chat-v3-0324z DeepSeek Chata5DeepSeek-V3 is the latest model from the DeepSeek team, building upon the instruction following and coding abilities of the previous versions. Pre-trained on nearly 15 trillion tokens, the reported evaluations reveal that the model outperforms other open-source models and rivals leading closed-source models.iúz0.135)r%é)Útzinfoé)Ú start_timeÚend_timez0.27)Ú constraintr%zdeepseek-reasonerz deepseek-r1zdeepseek-r1-0528z Deepseek R1z²DeepSeek R1 is here: Performance on par with OpenAI o1, but open-sourced and with fully open reasoning tokens. It's 671B parameters in size, with 37B active in an inference pass.z2.19)r r"r4r5rSr6r8r9Ú fireworksÚ Fireworkszhttps://api\.fireworks\.aizhttps://fireworks.ai/pricingzaccounts/fireworks/models/z*accounts/fireworks/models/deepseek-r1-0528zDeepSeek R1 0528aJThe updated DeepSeek-R1-0528 model delivers major improvements in reasoning, inference, and accuracy through enhanced post-training optimization and greater computational resources. It now performs at a level approaching top-tier models like O3 and Gemini 2.5 Pro, with notable gains in complex tasks such as math and programming.iqzdeepseek-v3-0324z*accounts/fireworks/models/deepseek-v3-0324zDeepseek V3 03-24zA strong Mixture-of-Experts (MoE) language model with 671B total parameters with 37B activated for each token from Deepseek. Updated checkpoint.zUdocs give just one price - "Pricing Per 1M Tokens", we assume that's input and outputz0.9zgemma-3-27b-itz(accounts/fireworks/models/gemma-3-27b-itzGemma 3 27B Instructi¸ÿ)r r!r"r$rSr%zllama-v3p1-8b-instructz0accounts/fireworks/models/llama-v3p1-8b-instructzLlama 3.1 8B Instructa…The Meta Llama 3.1 collection of multilingual large language models (LLMs) is a collection of pretrained and instruction tuned generative models in 8B, 70B and 405B sizes. The Llama 3.1 instruction tuned text only models (8B, 70B, 405B) are optimized for multilingual dialogue use cases and outperform many of the available open source and closed chat models on common industry benchmarks.zllama4-maverick-instruct-basicz8accounts/fireworks/models/llama4-maverick-instruct-basicz!Llama 4 Maverick Instruct (Basic)z0.22z0.88zqwen2p5-vl-72b-instructz1accounts/fireworks/models/qwen2p5-vl-72b-instructzQwen2.5-VL 72B InstructzLatest Qwen's VLM modelzqwen3-235b-a22bz)accounts/fireworks/models/qwen3-235b-a22bzQwen3 235B-A22Ba‚Qwen3 is the latest evolution in the Qwen LLM series, featuring both dense and MoE models with major advancements in reasoning, agent capabilities, multilingual support, and instruction following. It uniquely allows seamless switching between "thinking" (for complex logic, math, coding) and "non-thinking" modes (for fast, general dialogue), delivering strong performance across tasks.)r r"r4r5r6r8r9ÚgoogleÚGooglezhttps://(.*\.)?googleapis\.comz-https://ai.google.dev/gemini-api/docs/pricingz8https://cloud.google.com/vertex-ai/generative-ai/pricingÚgeminiÚvertexÚ usageMetadataÚpromptTokenCountÚcacheTokensDetailsz array-matchÚmodalityÚTEXT)ÚtypeÚfieldr!Ú tokenCountÚAUDIOÚcache_audio_read_tokensÚpromptTokensDetailsÚinput_audio_tokensÚcandidatesTokenCountÚthoughtsTokenCountÚ modelVersionzgemini-1.0-pro-vision-001zgemini 1.0 pro visionz¯Google's first-generation advanced multimodal model that can understand text, code, and images. It provides strong reasoning capabilities and follows instructions effectively.zWI can't find anything about this model or it's pricing, so trusting the original sourcez0.375zgemini-1.5-flashzgemini 1.5 flashzËA faster, more cost-effective variant of Gemini 1.5 that maintains strong capabilities while optimizing for performance and cost efficiency. Suitable for production deployments requiring high throughput.z0.01875zgemini-1.5-prozgemini 1.5 ProzÃGoogle's most capable multimodal model with an extremely long context window of up to 1 million tokens. It excels at complex reasoning, long-form content processing, and multimodal understanding.zgemini-2.0-flashzgemini-2.0-flash-lite-001zgemini-2.0-flash-001zgemini-2.0-flash-exp:freezgemini 2.0 flashzÁThe newest generation of Google's Gemini models, featuring improved reasoning, instruction following, and factual accuracy, with the Flash variant optimized for cost-efficiency and performance.z0.175z0.7)rr)rrKzgemini-2.0-flash-litezgemini 2.0 flash litezÆA lighter, more cost-effective version of Gemini 2.0 Flash, designed for applications requiring high efficiency while maintaining good performance. Ideal for high-volume, cost-sensitive deployments.zgemini-2.5-flashzgemini-2.5-flash-latestz gemini-2.5-flash-preview-09-2025zGemini 2.5 FlashaGemini 2.5 Flash is Google's state-of-the-art workhorse model, specifically designed for advanced reasoning, coding, mathematics, and scientific tasks. It includes built-in "thinking" capabilities, enabling it to provide responses with greater accuracy and nuanced context handling.)rr)rrKrMzgemini-2.5-flash-imagezgemini-2.5-flash-image-previewzGemini 2.5 Flash Imagez´Google's specialized image generation model optimized for fast, high-quality image generation. Outputs images at 1024x1024 resolution, with each image consuming 1290 output tokens.zÝSee https://ai.google.dev/gemini-api/docs/pricing#gemini-2.5-flash-image. Image output is priced at $30 per 1M tokens, with each 1024x1024 image = 1290 tokens = $0.039/image. Cache pricing is not available for this model.zgemini-2.5-flash-litezgemini-2.5-flash-lite-previewzGemini 2.5 Flash LiteaÐGemini 2.5 Flash-Lite is a lightweight reasoning model in the Gemini 2.5 family, optimized for ultra-low latency and cost efficiency. It offers improved throughput, faster token generation, and better performance across common benchmarks compared to earlier Flash models. By default, "thinking" (i.e. multi-pass reasoning) is disabled to prioritize speed, but developers can enable it via the Reasoning API parameter to selectively trade off cost for intelligence.z0.01zgemini-2.5-flash-previewzgemini-2.5-flash-preview-05-20zgemini-2.5-flash-preview-04-17z'gemini-2.5-flash-preview-05-20:thinkingz!gemini-2.5-flash-preview:thinkingzGemini 2.5 Flash Preview 05-20a.Gemini 2.5 Flash May 20th Checkpoint is Google's state-of-the-art workhorse model, specifically designed for advanced reasoning, coding, mathematics, and scientific tasks. It includes built-in "thinking" capabilities, enabling it to provide responses with greater accuracy and nuanced context handling.zZfrom https://cloud.google.com/vertex-ai/generative-ai/pricing should be retired 2025-07-15)r r!r"r#rSr%zgemini-2.5-prozGemini 2.5 ProaÚGemini 2.5 Pro is Google's state-of-the-art AI model designed for advanced reasoning, coding, mathematics, and scientific tasks. It employs "thinking" capabilities, enabling it to reason through responses with enhanced accuracy and nuanced context handling. Gemini 2.5 Pro achieves top-tier performance on multiple benchmarks, including first-place positioning on the LMArena leaderboard, reflecting superior human-preference alignment and complex problem-solving abilities.z@See https://ai.google.dev/gemini-api/docs/pricing#gemini-2.5-prozgemini-embedding-001zgemini-flash-1.5zGemini 1.5 Flasha8Gemini 1.5 Flash is a foundation model that performs well at a variety of multimodal tasks such as visual understanding, classification, summarization, and creating content from image, audio and video. It's adept at processing visual and text inputs such as photographs, documents, infographics, and screenshots.zBSee https://ai.google.dev/gemini-api/docs/pricing#gemini-1.5-flashzgemini-flash-1.5-8bzESee https://ai.google.dev/gemini-api/docs/pricing#gemini-1.5-flash-8bzgemini-live-2.5-flash-previewz%gemini-2.5-flash-native-audio-previewÚ12)rrrKrLz gemini-prozgemini-1.0-prozgemini 1.0 prozOI can't find anything so trusting these prices, not sure the model still existszgemini-pro-1.5zGemini 1.5 ProzVGoogle's latest multimodal model, supports image and video[0] in text or chat prompts.i€„z@See https://ai.google.dev/gemini-api/docs/pricing#gemini-1.5-proz0.3125z0.625zgemma-3zgemma-3-zGemma 3 (free)zgLightweight, state-of the art, open model built from the same technology that powers our Gemini models.zgemma-3nzGemma 3n (free)zlOur open model built for efficient performance on everyday devices like mobile phones, laptops, and tablets.ÚgroqÚGroqzhttps://api\.groq\.comzhttps://groq.com/pricing/zdeepseek-r1-distill-llama-70bzDeepSeek R1 Distill Llama 70Bz0.99)r r!r"r$r%z gemma-7b-itz gemma2-9b-itz gemma2-9bz Gemma 2 9B 8k)r r!r"r%zllama-3.1-405b-reasoningz0.59z0.79zllama-3.1-70b-versatilezllama-3.1-8b-instantzLlama 3.1 8B Instant 128kzllama-3.2-11b-text-previewz0.18zllama-3.2-11b-vision-previewzllama-3.2-1b-previewz0.04zllama-3.2-3b-previewzllama-3.2-90b-text-previewzllama-3.2-90b-vision-previewzllama-3.3-70b-specdeczllama-3.3-70b-versatilezLlama 3.3 70B Versatile 128kzllama-guard-3-8bzllama2-70b-4096zllama3-70b-8192zllama3-8b-8192z%llama3-groq-70b-8192-tool-use-previewz0.89z$llama3-groq-8b-8192-tool-use-previewz0.19z-meta-llama/llama-4-maverick-17b-128e-instructzLlama 4 Maverick 17B 128Ez)meta-llama/llama-4-scout-17b-16e-instructzLlama 4 Scout (17Bx16E) 128kz0.11z0.34zmeta-llama/llama-guard-4-12bzLlama Guard 4 12Bzmistral-saba-24bzmixtral-8x7b-32768zmoonshotai/kimi-k2-instructz moonshotai/kimi-k2-instruct-0905zKimi K2 1T 128kzopenai/gpt-oss-120bzopenai/gpt-oss-safeguard-20bz“GPT-OSS 120B is OpenAI's flagship open source model, built on a Mixture-of-Experts (MoE) architecture with 120 billion parameters and 128 experts. )r r!r#r$r%zopenai/gpt-oss-20bzGPT-OSS 20B is OpenAI's flagship open source model, built on a Mixture-of-Experts (MoE) architecture with 20 billion parameters and 32 experts. zqwen/qwen3-32bzQwen3 32B 131kz0.29)r r"r4r5r8r9ÚmistralÚMistralzhttps://api\.mistral\.aiz&https://mistral.ai/pricing#api-pricingz(?:mi|code|dev|magi|mini)stral)ÚregexÚ codestralzcodestral-latestzcodestral-2501Ú Codestralz³Mistral's cutting-edge language model for coding. Codestral specializes in low-latency, high-frequency tasks such as fill-in-the-middle (FIM), code correction and test generation.zdevstral-smallzDevstral SmallaSDevstral-Small-2505 is a 24B parameter agentic LLM fine-tuned from Mistral-Small-3.1, jointly developed by Mistral AI and All Hands AI for advanced software engineering tasks. It is optimized for codebase exploration, multi-file editing, and integration into coding agents, achieving state-of-the-art results on SWE-Bench Verified (46.8%).z0.12zdevstral-small:freezDevstral Small (free)zmagistral-mediumzMagistral MediumubMagistral is Mistral's first reasoning model. It is ideal for general purpose use requiring longer thought processing and better accuracy than with non-reasoning LLMs. From legal research and financial forecasting to software development and creative storytelling — this model solves multi-step challenges where transparency and precision are critical.zmagistral-smallzmagistral-small-zMagistral Smalla8Magistral Small is a 24B parameter instruction-tuned model based on Mistral-Small-3.1 (2503), enhanced through supervised fine-tuning on traces from Magistral Medium and further refined via reinforcement learning. It is optimized for reasoning and supports a wide multilingual range, including over 20 languages.z ministral-3bz Ministral 3BaYMinistral 3B is a 3B parameter model optimized for on-device and edge computing. It excels in knowledge, commonsense reasoning, and function-calling, outperforming larger models like Mistral 7B on most benchmarks. Supporting up to 128k context length, it's ideal for orchestrating agentic workflows and specialist tasks with efficient inference.z ministral-8bzMinistral 8B 24.10ajMinistral 8B is an 8B parameter model featuring a unique interleaved sliding-window attention pattern for faster, memory-efficient inference. Designed for edge use cases, it supports up to 128k context length and excels in knowledge and reasoning tasks. It outperforms peers in the sub-10B category, making it perfect for low-latency, privacy-first applications.z mistral-7bzopen-mistral-7bz Mistral 7Bz mistral-embedz mistral-largezmistral-large-latestzmistral-large-2407zmistral-large-2411z Mistral Largez×This is Mistral AI's flagship model, Mistral Large 2 (version `mistral-large-2407`). It's a proprietary weights-available model and excels at reasoning, code, JSON, chat, and more. Read the launch announcement here.zmistral-medium-3zmistral-mediumzMistral Medium 3uwMistral Medium 3 is a high-performance enterprise-grade language model designed to deliver frontier-level capabilities at significantly reduced operational cost. It balances state-of-the-art reasoning and multimodal performance with 8× lower cost compared to traditional large models, making it suitable for scalable deployments across professional and industrial use cases.z mistral-nemozopen-mistral-nemoz Mistral NeMozeA 12B parameter model with a 128k token context length built by Mistral in collaboration with NVIDIA.zmistral-nemo:freezMistral Nemo (free)z mistral-sabazmistral-saba-latestz Mistral SabauÈMistral Saba is a 24B-parameter language model specifically designed for the Middle East and South Asia, delivering accurate and contextually relevant responses while maintaining efficient performance. Trained on curated regional datasets, it supports multiple Indian-origin languages—including Tamil and Malayalam—alongside Arabic. This makes it a versatile option for a range of regional and multilingual applications. Read more at the blog post herez mistral-smallz Mistral Smalla½With 22 billion parameters, Mistral Small v24.09 offers a convenient mid-point between (Mistral NeMo 12B)[/mistralai/mistral-nemo] and (Mistral Large 2)[/mistralai/mistral-large], providing a cost-effective solution that can be deployed across various platforms and environments. It has better reasoning, exhibits more capabilities, can produce and reason about code, and is multiligual, supporting English, French, German, Italian, and Spanish.zmistral-small-24b-instruct-2501zMistral Small 3zûMistral Small 3 is a 24B-parameter language model optimized for low-latency performance across common AI tasks. Released under the Apache 2.0 license, it features both pre-trained and instruction-tuned versions designed for efficient local deployment.z>Can't find pricing on this model, so just trusting open routerz$mistral-small-24b-instruct-2501:freezMistral Small 3 (free)zmistral-small-latestzMistral Small 3.2z+SOTA. Multimodal. Multilingual. Apache 2.0.z mistral-tinyz Mistral TinyzWNote: This model is being deprecated. Recommended replacement is the newer Ministral 8Bzmixtral-8x22b-instructzMixtral 8x22B Instructa3Mistral's official instruct fine-tuned version of Mixtral 8x22B. It uses 39B active parameters out of 141B, offering unparalleled cost efficiency for its size. Its strengths include: - strong math, coding, and reasoning - large context length (64k) - fluency in English, French, Italian, German, and Spanishz mixtral-8x7bzopen-mixtral-8x7bz Mixtral 8x7Bz pixtral-12bzpixtral-12b-latestz Pixtral 12Bz›The first multi-modal, text+image-to-text model from Mistral AI. Its weights were launched via torrent: https://x.com/mistralai/status/1833758285167722836.z pixtral-largezpixtral-large-latestzpixtral-large-2411zPixtral Large 2411z§Pixtral Large is a 124B parameter, open-weight, multimodal model built on top of Mistral Large 2. The model is able to understand documents, charts and natural images.ÚnovitaÚNovitazhttps://api\.novita\.aizhttps://novita.ai/pricingzSao10K/L3-8B-Stheno-v3.2z+cognitivecomputations/dolphin-mixtral-8x22bzdeepseek/deepseek-r1z&deepseek/deepseek-r1-distill-llama-70bz%deepseek/deepseek-r1-distill-llama-8bz%deepseek/deepseek-r1-distill-qwen-14bz%deepseek/deepseek-r1-distill-qwen-32bzdeepseek/deepseek_v3zgoogle/gemma-2-9b-itzgryphe/mythomax-l2-13bz0.09zjondurbin/airoboros-l2-70bzmeta-llama/llama-3-70b-instructz0.51z0.74zmeta-llama/llama-3-8b-instructz!meta-llama/llama-3.1-70b-instructz0.39z meta-llama/llama-3.1-8b-instructz$meta-llama/llama-3.1-8b-instruct-maxz%meta-llama/llama-3.1-8b-instruct-bf16z(meta-llama/llama-3.2-11b-vision-instructz meta-llama/llama-3.2-1b-instructz meta-llama/llama-3.2-3b-instructz!meta-llama/llama-3.3-70b-instructzmicrosoft/wizardlm-2-8x22bz0.62zmistralai/mistral-7b-instructz0.059zmistralai/mistral-nemoz0.17z$nousresearch/hermes-2-pro-llama-3-8bz#nousresearch/nous-hermes-llama2-13bzopenchat/openchat-7bzqwen/qwen-2-7b-instructz0.054zqwen/qwen-2-vl-72b-instructzqwen/qwen-2.5-72b-instructz0.38zsao10k/l3-70b-euryale-v2.1z1.48zsao10k/l3-8b-lunariszsao10k/l31-70b-euryale-v2.2z!sophosympatheia/midnight-rose-70bz!teknium/openhermes-2.5-mistral-7bÚopenaiÚOpenAIzhttps://api\.openai\.com)z(https://platform.openai.com/docs/pricingzhttps://openai.com/api/pricing/z'https://platform.openai.com/docs/modelszDhttps://help.openai.com/en/articles/7127956-how-much-does-gpt-4-costzgpt-z^o[134]Úinput_tokens_detailsÚ responsesÚ embeddingsz text-ada-001zchatgpt-4o-latestz ChatGPT-4ozöOpenAI ChatGPT 4o is continually updated by OpenAI to point to the current version of GPT-4o used by ChatGPT. It therefore differs slightly from the API version of GPT-4o in that it has additional RLHF. It is intended for research and evaluation.z codex-minizcodex-mini-latestz Codex Miniz–codex-mini-latest is a fine-tuned version of o4-mini specifically for use in Codex CLI. For direct use in the API, we recommend starting with gpt-4.1.ztext-davinci-001zft:gpt-3.5-turbo-zft:gpt-3.5-turbozGPT-3.5 Turbo fine tuned.z ft:gpt-4ozft:gpt-4o-2024-zGPT-4o fine tuned.zft:gpt-4o-minizft:gpt-4o-mini-2024-zGPT-4o Mini fine tuned.z gpt-3.5-0301z gpt 3.5 turbozsΰ    ÿ  ø óÿúþÿüõþÿüõüÿüóüú üú üú üÿüóüüüÿüóþÿüõûÿ üòþÿÿúõü¨èxýýýýðû  þùÿ ýÿû ÿû ÿûèï4üÿ öÿ  ýóýÿ ÷îüÿ öã)ûÿ õ×6ýÊ;ýÅ@ýÀEý»Jýÿ÷¶Uÿû«\ýÿ÷¤gý™lý”qývýÿ÷Šýþüÿöùýíýèÿÿûã&ÿÿûÚ/ýÑ4ÿýÌ;ÿýÅBýÿûö¾Túý¬`úý lÿú”vÿúŠÿúÿ  ýõ ýð ûë    üÿöä(ÿýØ/ýÑ4ÿýÌ;ÿýÅB û¾I û·P û°W û©^ û¢e û›l û”s ûz ý† ý ûûúþ  þùÿ    ýÿó   ýÿó   ýÿó   ýÿóÒëT    þùÿ  û ûûÿù ûáî: ÿÿø óÿÿÿÿþÿûúø   ýÿÿÿþÿûúôæèQ ÿÿø óÿ ú ù  ú ù  ú ù  úÌéUþ    ýÿýù ýù ýù â Ý%   ÿ  ø óÚ7üü üü üü üü üý ÿüú ÿüú  ù  ÿûú þú     üÿÿúó ú   ýÿûõ  þÿô  þÿûõ     ûÿ ò ÿûú  ý ÿûú ÿûù  þÿüøù  ÿûùûû˺þùÿ  û ýü ý ý ü ý ý ý ý ý ý ý ü ý ý ý ý ý ý û ü û ý ý  þÿö   þÿÿö ÿû  üÔðD  þùÿ û û ûû û û ûü ý    üÿô ûû ûû û ú û û û ûÿúûÿùÞî? ýÿ ýú  ýõ ýð ýë ýæ ýá$ ýÜ) ý×. ýÒ3 ýÍ8 ýÈ= ýÃB ý¾G  þÿø¹Q ý¯V ýª[ ý¥` ý e ý›j ý–o ý‘t ýŒy ý‡~ ý‚ ýü  ý÷ ýò ýí ýè ýã" ýÞ' ýÙ, ýÔû8 ÿÿÿõ ð ÿ ûö üá&ýÿýú  ûõÿûîýå ýà% üÛ+ üÕ1 üÏ7ýÉ<ýÿôÄJü¶Pü°V üÿóªeþÿõ›r üÿóŽýÿóþúÿ ñï"ÿøÞ,úÔ4ÿÿøÌ@ÿÿøÀL ÿû´U üÿÿó«f úšn   üÿÿó’ üûú úýó ýç ýâ#úýÝ/  þÿöÑ;ÿûÅB û¾I   ûÿ ÿò·[ ü¥a üŸgÿú™qÿú{ü…     ûÿ ÿòþÿúúì# úüÝ0 üÐ6    üÿóÊEÿú»Oû±Vÿÿÿþúûªiÿû—p   ýÿÿõû   ýÿÿõù  þÿøê ýà% ýÛ* úÖ2 úÎ:ýÿ ôÆÎ| ýÿ ýú  ýõ ýð ýë ýæ ýá$ ýÜ) ü×/ üÑ5 ýË: ýÆ? ýÁD ü¼J ý¶O ý±Tÿ¬W ý©\ ý¤a ýŸf ýšk ý•p ýu ý‹z      úÿ ô†   þÿø÷  þÿøí  þÿøã'    üÿöÙ3    üÿöÍ?   ýÿ÷ÁJ ü¶P ü°V ýª[ ý¥` ü f üšl     úÿ ó”{üü…üüùÿüúî   üÿüõá1   üÿüõÏC  ýÿüö½Tüü¬_üü¡j ü–p üv üŠ| ÿü„ ýû  ýö ýñ ýì ýç ýâ#ÿûÝ* ýÖ/  þÿøÑ9 ýÇ> ýÂC ü½I ü·Oü±U ü«[ÿú¥c üi ü—o ü‘u ü‹{ ü… üþ üø üò üì üæ  üà& üÚ, üÔ2 üÎ8 üÈ> üÂD ü¼J ü¶P ü°V üª\ ü¤b üžh ü˜n ü’t üŒz ü† ûÿ ýø  ýó ýî ûé ýâ# ýÝ( ýØ- ýÓ2 ýÎ7 ýÉ< ýÄA ý¿F ýºK ýµP ý°U ý«Z ý¦_ ü¡e ü›k ü•q üw ü‰} üƒ üü  ýö üñ üë üå! ýß& ýÚ+ ýÕ0 ýÐ5 üË; üüÅF üºL ü´Rÿüú®_ ü¡e  þÿü÷›u üü‹ üüÿ       úÿ üóô  üà& üüÚ1 üüÏ< üÄB ü¾H ü¸N ü²T ü¬Z ü¦` ü f üšl ü”r üŽx üˆ~ ü‚ üû  üõ üï üé üã# ýÝ( ýØ- ýÓ2 ýÎ7 ýÉ<  þÿ÷ÄG üü¹R ü®X ý¨] üü£h ý˜m ý“r ýŽw ý‰| ý„ ýþÿûù ýò ýí ýè ýã" ýÞ' ýÙ, ýÔ1 ýÏ6 ýÊ; ýÅ@ ýÀE ý»J  þÿø¶T  þÿø¬^ü¢düœjü–p üvüŠ|ü„üý ü÷üñÿüëÿüã% ÿüÛ- ýÿÿöÓ;üÅAÿü¿I ü·O ü±U ü«[ ü¥a üŸgü™mÿü“u ü‹{ ü… ýþ üù  üó üí ýç ýâ# üÝ) ü×/ ýÑ4 ýÌ9 üÇ? üÁE ü»K üµQ ý¯V üª\ ü¤b üžh ü˜n ü’t üŒz ý† ü üú  üô ýî ýé ýä! üß' üÙ- üÓ3 üÍ9 üÇ? üÁE ü»K üµQ ü¯W ü©] ü£c üi ü—o ü‘u ü‹{ ü… üþ üø üò üì üæ  üà& üÚ, üÔ2 üÎ8 üÈ> üÂD ü¼J ü¶P ü°V üª\ ü¤b üžh ü˜n ü’t üŒz ü† üÿ  þÿ÷ù üî üè üâ$ üÜ* üÖ0 ýÐ5 üË; ýÅ@ ýÀE ý»J ý¶O ý±T ý¬Y ý§^ ý¢c ýh ý˜m ý“r ýŽw ý‰| ý„ ýþ ýù  ýô ýï ýê ýå  ýà% ýÛ* ýÖ/ ýÑ4 ýÌ9 ýÇ> ýÂC ý½H ý¸M ý³R ý®W ý©\ ü¤b üžh ü˜n ü’t ýŒy ü‡ üÿúú üò üì üæ    ýÿöà, üÔ2 üÎ8 üÈ> üÂD ü¼J ü¶P ü°V üª\ ü¤b üžh ü˜n ü’t ýŒy ý‡~ ý‚ ýü  ý÷  þÿøò ýè ýã" ýÞ'   ýÿ÷Ù2 ýÎ7 ýÉ< ýÄA ý¿F ýºK ýµP ý°U ý«Z ý¦_ ý¡d ýœi ý—n ý’s ýx üˆ~ ü‚ üû  üõ üï ýé ýä! üß' üÙ- ýÓ2  þÿøÎ< ýÄA ý¿F ýºK üµQ ý¯V üª\ ý¤a ýŸf ýšk ý•p ýu ý‹zü†üÿüù  ýó ýî ýé ýä! ýß&   ýÿöÚ2ÿüÎ: üÆ@ üÀFÿüºN ü²Tÿü¬\ ý¤a ýŸf ýškÿû•r ýŽw ý‰| ý„ ýþýù  ýôÿûïÿûè ýá$ ýÜ) ý×. ýÒ3    üÿöÍ? ýÁD   ýÿ÷¼O ý±T   ýÿ÷¬_ÿû¡f ýšk ý•pýuý‹z ý† ü ýú  ýõ ýð ýë ýæ ýá$ ýÜ) ý×. üÒ4 üÌ: üÆ@ üÀF üºL ü´R ü®X ü¨^ ü¢d üœj ý–o ü‘u ü‹{ ü… üþ üø üò üì ÿüæ" ÿüÞ* ÿüÖ2 üÎ8 üÈ> ýÂC ý½H ý¸M ý³R ý®W ý©\ ý¤a ýŸf ýšk ý•p ýu ý‹z ý† ý ýû  ýö ýñ ýì ýç ýâ# ýÝ( ýØ- ýÓ2 ýÎ7É8 üÈ> üÂD ü¼J ü¶P ü°V üª\ ü¤b üžh ü˜n ü’t üŒz ü† üÿ üù  üó üí üç üá%ÿÛ( üØ. ýÒ3 üÍ9 ýÇ> üÂD ü¼J ý¶O ý±T ý¬Y ý§^ ý¢c üi ý—n ý’s ýx üˆ~ ü‚ üû  üõ üï üé üã# ýÝ( ýØ- üÓ3 üÍ9 ýÇ> ýÂC ý½H ý¸M ý³Rÿ®Uÿ«X ü¨^ ý¢c ýh ü˜n ü’t üŒz ü† üÿ üù  ýó ýî ýé ýä! ýß& ýÚ+ ýÕ0 üÐ6 üÊ< üÄûI û û û û û û û ûÎúA ýÿ ýú  ýõ ýð ýë ýæ ýá$ ýÜ) ý×. ýÒ3ÿûÍ:ÿûÆAÿû¿Hÿû¸O ý±Tÿû¬[ ý¥` ý e ý›j   ýÿ÷–u ý‹z ý† ý ýû  ýö ýñýìýç ýâ# ýÝ( ýØ- ýÓ2 ýÎ7 ýÉ< ýÄA ý¿F ýºK ýµP ý°U ý«Z ý¦_ ý¡d ýœi ý—n ý’s ýx ýˆ} ýƒ ýý ýø  ýó ýî ýé ýä! ýß& ýÚ+ ýÕ0 ýÐ5 ýË: ýÆ? ýÁD ý¼I ý·N ý²S ý­X ý¨] ý£b ýžg ý™l ý”q ýv ýŠú  ÿÿø óÿ   ýÿô   ýÿô   ýÿô   ýÿô   ýÿÿô   ýÿô   ýÿô©èž