All promotions
GuidesMCP

Models

208 models · 864 offers · prices read Oct 23, 09:51 UTC

Columns
ModelReleasedIntel.InputOutputCacheProviderstok/sContextPriciest hostGapVerified
Muse Spark Open openMetaOct 15155$1.25$4.25$0.1552631M$5.50 at Modal+0%Oct 23
Nemotron 3.5 Super VL openNVIDIAOct 15—$0.10$0.40$0.0273881M$1.50 at Venice+200%Oct 23
Muse WonderMetaOct 12162$4.00$20.00$0.201591M——Oct 23
Claude Haiku 5.5AnthropicOct 7159$0.10$0.50$0.011—1M——Oct 7
Mistral Large 4Mistral AIOct 6150$1.36$4.18$0.141461M——Oct 23
Gemini 4 ArgonGoogleSep 30166$2.00$10.00$0.101—1M——Oct 17
GPT-6.1 SolOpenAISep 29168$2.00$10.00$0.101—1.05M——Sep 29
Ling 3.1 Flash open −75% at DeepInfrainclusionAISep 29—$0.075$0.23$0.01527256K$1.20 at Novita AI+300%Oct 23
Claude Sonnet 5.5AnthropicSep 28166$2.00$10.00$0.101—1M——Oct 17
Qwen3.8 Max PrimeQwenSep 23—$4.00$12.00$0.501671M——Oct 23
Claude Opus 5.5AnthropicSep 22169$4.00$20.00$0.201—1M——Sep 24
Command A+ openCohereSep 22—$0.30$1.50$0.151129192K——Oct 23
GPT-6 LunaOpenAISep 22157$0.10$0.50$0.011—1.05M——Sep 28
GPT-6 SolOpenAISep 22164$2.00$10.00$0.201—1.05M——Sep 28
Grok 4.7xAISep 21154$2.00$6.00$0.501—500K——Sep 24
MiMo-V2.6-Flash open −15% at GMICloudXiaomi MiMoSep 21—$0.14$0.14$0.002691741.05M$0.53 at Venice+87%Oct 23
MiMo-V2.6-Pro open −10% at DeepInfraXiaomi MiMoSep 21—$0.39$0.78$0.00325891.05M$13.05 at Xiaomi+1015%Oct 23
GLM 5.3 FlashX openGLMSep 18—$0.37$1.25$0.091751M——Oct 23
Qwen3.8 Omni FlashAlibabaSep 17—$0.15$0.47$0.0161—1M——Sep 18
Step 5 PreviewStepFunSep 16—$1.00$2.70$0.1021041M$3.70 at Novita AI+0%Oct 23
DeepSeek V4.1 Flash open −51% at StreamLakeDeepSeekSep 10156$0.099$0.37$0.002292711M$1.65 at Alibaba+253%Oct 2
Ling 3.0 Flash Sante (inclusionAI)inclusionAISep 4—$0.075$0.22$0.0151—256K——Oct 23
Ling 3.0 Flash VL (inclusionAI) openinclusionAISep 4—$0.06$0.18$0.012243256K$0.29 at Novita AI+23%Oct 23
GPT-6 AstraOpenAISep 3169$10.00$50.00$1.001—1M——Oct 17
Ling 3.0 Flash Fin (inclusionAI) openinclusionAISep 3—$0.075$0.22$0.0151—256K——Oct 23
Gemini 3.8 FlashGoogleSep 2157$0.75$3.75$0.0751—1M——Oct 17
Muse Spark 1.3MetaSep 2157$1.25$4.25$0.1512071M——Oct 23
Claude Fable 5.1AnthropicSep 1166$10.00$50.00$0.253—1M$60.00 at Google+0%Sep 4
Granite 4.2 8B openIBMAug 31—$0.10$0.15$0.0152107128K$0.31 at DeepInfra+24%Oct 23
Hunyuan 4 PreviewTencentAug 28—$0.83$2.50$0.0421—1M——Oct 23
GLM-5.3-Flash open −71% at StreamLakeZ.aiAug 26153$0.0435$0.14$0.0075291401M$2.48 at Reka+1216%Oct 23
Qwen3.8 FlashQwenAug 26—$0.15$0.47$0.0163—1M$0.62 at Novita AI+0%Sep 5
DeepSeek V4 Flash Vision Exp open −51% at DeepInfraDeepSeekAug 21—$0.22$0.65$0.006971231M$1.76 at GMICloud+104%Oct 23
GLM-5.3 open −50% at SiliconFlowZ.aiAug 14156$0.40$1.96$0.067321301M$6.96 at Amazon Bedrock+195%Oct 23
Qwen3.8 27B open −25% at DeepInfraAlibabaAug 14152$0.039$1.25$0.0039161011M$4.80 at Groq+272%Oct 23
Gemini 3.7 FlashGoogleAug 13158$0.75$3.75$0.0751—1M——Sep 4
Grok 4.6xAIAug 12157$2.00$6.00$0.502—250K$8.80 at Amazon Bedrock+10%Oct 17
Qwen3.8 2.4T A95B open −20% at VeniceAlibabaAug 12—$2.00$6.00$0.2071571M$8.00 at Venice+0%Oct 23
Nemotron 3.5 Lightning open −33% at Io NetNVIDIAAug 11—$0.0469$0.13$0.02354167256K$0.27 at Phala+49%Oct 23
Muse Glimmer 30B (Meta) openMetaAug 10—$0.30$1.20$0.04291128K$1.85 at Together AI+23%Oct 23
Qwen3.8 MaxQwenAug 3157$2.00$6.00$0.252331M$8.00 at Alibaba+0%Oct 23
Inkling Small openThinking MachinesJul 30150$0.45$1.20$0.1031911M$1.70 at Together AI+3%Oct 23
Claude Opus 5AnthropicJul 24164$5.00$25.00$0.503671M$30.00 at Google+0%Oct 23
Ling 3.0 Flash openinclusionAIJul 23—$0.06$0.18$0.012247256K$0.24 at DeepInfra+0%Oct 23
Gemini 3.5 Flash-LiteGoogleJul 21145$0.30$2.50$0.031—1M——Sep 4
Gemini 3.6 FlashGoogleJul 21155$0.75$3.75$0.0751—1M——Sep 3
Kimi K3 open −15% at PhalaMoonshotJul 16158$1.85$9.50$0.1817731M$20.70 at Alibaba+82%Oct 23
Inkling openThinking MachinesJul 15148$0.95$4.05$0.163951M$5.05 at Together AI+1%Oct 22
Qwen3.7 FlashQwenJul 15144$0.03$0.13$0.0061901M——Oct 23
GPT-5.6 LunaOpenAIJul 9156$0.20$1.20$0.0231361.05M$1.54 at Amazon Bedrock+10%Oct 23
GPT-5.6 SolOpenAIJul 9163$4.00$20.00$0.403831.05M$26.40 at Amazon Bedrock+10%Oct 23
GPT-5.6 TerraOpenAIJul 9160$2.00$12.00$0.203881.05M$15.40 at Amazon Bedrock+10%Oct 23
Grok 4.5xAIJul 8154$2.00$6.00$0.302—500K$8.80 at Amazon Bedrock+10%Sep 4
Hunyuan 3 openTencentJul 6—$0.13$0.53$0.0332—256K$0.72 at Novita AI+9%Oct 8
Claude Sonnet 5AnthropicJun 30157$2.00$10.00$0.203821M$12.00 at Google+0%Oct 23
GLM 5.2 open −56% at StreamLakeZ.aiJun 13152$0.56$1.80$0.067272231M$8.19 at Cloudflare+247%Oct 23
Kimi K2.7 Code open −25% at StreamLakeMoonshotJun 12150$0.71$3.00$0.141598256K$4.95 at Phala+33%Oct 23
Claude Fable 5AnthropicJun 9163$10.00$50.00$1.003621M$60.00 at Google+0%Oct 23
Nemotron 3 Ultra openNVIDIAJun 4146$0.50$2.20$0.10382256K$3.75 at Venice+39%Oct 23
Nemotron 3.5 Content Safety openNVIDIAJun 4—$0.20$0.20—1224128K——Oct 23
Qwen3.7 PlusQwenJun 2147$0.40$1.60$0.081481M——Oct 23
MiniMax-M3 open −50% at StreamLakeMiniMaxJun 1147$0.23$0.96$0.048122651M$3.00 at SambaNova+152%Oct 23
Step 3.7 FlashStepFunMay 29—$0.20$1.15$0.041—256K——Oct 23
Qwen3.7 MaxQwenMay 21154$1.25$3.75$0.252671M$10.00 at Alibaba+100%Oct 23
Mistral Medium 3.5 openMistral AIApr 30141$1.50$7.50—1142256K——Oct 23
Qwen3.6 FlashQwenApr 27143$0.25$1.50$0.0251661M——Oct 23
DeepSeek V4 Flash 0423 open −50% at StreamLakeDeepSeekApr 24146$0.07$0.14$0.014121061M$1.76 at Cloudflare+738%Oct 23
DeepSeek V4 Flash 0731 open −70% at StreamLakeDeepSeekApr 24155$0.07$0.14$0.0042252451M$1.76 at Cloudflare+738%Oct 23
DeepSeek V4 Pro 0423 open −45% at StreamLakeDeepSeekApr 24149$0.90$1.50$0.0796151291M$5.74 at Azure OpenAI+139%Oct 23
DeepSeek V4 Pro 0813 open −50% at DeepSeekDeepSeekApr 24156$0.50$2.00$0.022231391M$6.60 at Venice+164%Oct 23
MiMo-V2.5 −5% at DeepInfraMiMoApr 22—$0.13$0.27$0.00275141.05M$2.40 at Venice+502%Sep 29
MiMo-V2.5-Pro −61% at DeepInfraMiMoApr 22—$0.30$0.61$0.002871101.05M$3.80 at DigitalOcean+316%Oct 23
Qwen3.6 27B openAlibabaApr 22146$0.25$2.00$0.025951256K$4.20 at Novita AI+87%Oct 23
Kimi K2.6 open −37% at StreamLakeMoonshotApr 21151$0.45$1.95$0.0451894256K$5.69 at Phala+137%Oct 23
Qwen3.6 Max PreviewQwenApr 20149$1.30$7.80$0.131—256K——Oct 23
Qwen3.6 35B A3B openAlibabaApr 17144$0.10$0.90$0.0512176256K$2.04 at SiliconFlow+104%Oct 23
GLM 5.1 open −31% at StreamLakeZ.aiApr 7150$0.97$3.04$0.09816120200K$5.81 at Venice+45%Oct 23
Gemma 4 26B A4B open −15% at NextBitGoogleApr 2142$0.06$0.20$0.0351264256K$0.75 at Google+188%Sep 30
Gemma 4 31B open −5% at Io NetGoogleApr 2143$0.08$0.30$0.01214168256K$1.75 at SiliconFlow+361%Sep 29
Qwen3.6 PlusQwenApr 2147$0.50$3.00$0.051461M——Oct 23
GLM 5V Turbo openGLMApr 1—$1.20$4.00$0.24235198K$5.20 at Novita AI+0%Oct 23
Reka EdgeRekaMar 20—$0.10$0.10—13516K——Oct 23
MiniMax M2.7 open −30% at GMICloudMiniMaxMar 18146$0.21$0.84$0.0427264200K$3.00 at SambaNova+186%Oct 23
GLM 5 Turbo openGLMMar 16—$1.20$4.00$0.24121198K——Oct 23
Mistral Small 4 openMistral AIMar 16—$0.15$0.60—2—256K$0.94 at Venice+25%Sep 3
Nemotron 3 Super openNVIDIAMar 11—$0.085$0.40$0.0641211M$2.70 at Crusoe+457%Oct 23
GPT-5.3 CodexOpenAIFeb 24157$1.75$14.00$0.171—400K——Oct 17
Qwen3.5 122B A10B openAlibabaFeb 23—$0.26$2.08$0.30597256K$3.60 at Novita AI+54%Oct 23
Qwen3.5 9B openAlibabaFeb 23139$0.10$0.15—574256K$0.42 at Together AI+68%Oct 23
Qwen3.5-27B openAlibabaFeb 23—$0.25$2.00$0.026773256K$2.86 at Reka+27%Oct 23
Qwen3.5-35B-A3B openAlibabaFeb 23142$0.14$1.00$0.058127256K$2.25 at Novita AI+97%Oct 23
Qwen3.5-FlashQwenFeb 23144$0.10$0.40$0.011701M——Oct 23
Gemini 3.1 Pro PreviewGoogleFeb 19155$2.00$12.00$0.201—1M——Oct 17
Qwen3.5 Plus 2026-04-20QwenFeb 16146$0.40$2.40—1451M——Oct 23
Qwen3.5 397B A17B openAlibabaFeb 15147$0.45$3.00$0.111052256K$5.25 at Venice+52%Oct 23
GLM 5 open −40% at GMICloudZ.aiFeb 12146$0.60$1.92$0.1210118200K$4.20 at Z.AI+67%Oct 23
MiniMax M2.5 open −10% at StreamLakeMiniMaxFeb 12147$0.27$0.95$0.027981200K$1.50 at GMICloud+23%Oct 23
Qwen3 Coder Next open −40% at StreamLakeAlibabaFeb 3—$0.12$0.80$0.036475256K$1.80 at Alibaba+96%Oct 23
Kimi K2.5 open −5% at VeniceMoonshotJan 27148$0.45$2.25$0.07867256K$3.86 at Venice+43%Sep 7
MiniMax M2-her openMiniMaxJan 23—$0.30$1.20$0.0314164K——Oct 23
Qwen3 Max ThinkingQwenJan 23—$0.78$3.90—1—256K——Oct 9
GLM 4.7 Flash openZ.aiJan 19—$0.06$0.40$0.01443198K$0.47 at Novita AI+2%Oct 19
DeepSeek V4.1 Flash 1017 open −51% at DeepInfraDeepSeekJan 1—$0.098$0.29$0.0029222681M$1.65 at Alibaba+321%Oct 23
MiniMax M3.1 FlashMiniMaxJan 1—$0.10$0.50$0.00542281M$0.75 at Together AI+25%Oct 23
Qwen3 VL 8B Thinking openAlibabaJan 1—$0.18$2.10—1—128K——Oct 9
MiniMax M2.1 openMiniMaxDec 23—$0.30$1.20$0.03226200K$1.50 at Minimax+0%Oct 8
GLM 4.7 open −27% at VeniceZ.aiDec 22144$0.40$1.75$0.08626200K$2.80 at Z.AI+30%Oct 23
Gemini 3 Flash PreviewGoogleDec 17152$0.50$3.00$0.051—1M——Oct 17
Nemotron 3 Nano 30B A3B openNVIDIADec 15—$0.05$0.20$0.0254173256K$0.30 at Nebius+20%Oct 7
Devstral 2 2512MistralDec 9—$0.40$2.00$0.04192256K——Oct 23
GLM 4.6V openZ.aiDec 8—$0.30$0.90$0.05210128K$1.20 at Z.AI+0%Oct 23
Ministral 3 14B 2512 openMistral AIDec 2—$0.20$0.20$0.02164256K——Oct 23
Ministral 3 3B 2512 openMistral AIDec 2—$0.10$0.10$0.011189128K——Oct 23
Ministral 3 8B 2512 openMistral AIDec 2—$0.15$0.15$0.0151116256K——Oct 23
Mistral Large 3 openMistral AIDec 2—$0.50$1.50—1—128K——Sep 3
Kimi K2 Thinking openMoonshotNov 6146$0.60$2.50$0.152—256K$3.10 at Novita AI+0%Oct 22
gpt-oss-safeguard-20b openOpenAIOct 29—$0.075$0.30$0.03751588128K——Oct 23
MiniMax M2 open −15% at MinimaxMiniMaxOct 27—$0.26$1.02$0.03325200K$1.50 at Novita AI+18%Oct 23
Qwen3 VL 32B InstructQwenOct 21—$0.10$0.42—1—128K——Oct 9
Claude Haiku 4.5AnthropicOct 15142$1.00$5.00$0.101—200K——Oct 17
Qwen3 VL 8B Instruct openAlibabaOct 14—$0.12$0.46$0.12263256K$1.00 at Parasail+75%Oct 9
Qwen3 VL 30B A3B Thinking openAlibabaOct 6—$0.29$1.00—228256K$2.60 at Alibaba+102%Oct 23
Qwen3 VL 30B A3B Instruct openAlibabaOct 4—$0.15$0.60—424256K$1.29 at SiliconFlow+72%Oct 23
Granite 4.0 MicroIBMOct 2—$0.017$0.11—135131K——Oct 23
GLM 4.6 openZ.aiSep 30—$0.43$1.75$0.08532200K$2.80 at Z.AI+28%Oct 23
Claude Sonnet 4.5AnthropicSep 29147$3.00$15.00$0.301—200K——Oct 17
DeepSeek V3.2 open −28% at GMICloudDeepSeekSep 29146$0.21$0.31$0.02151461160K$7.50 at SambaNova+1347%Oct 23
DeepSeek V3.2 Exp openDeepSeekSep 29145$0.13$0.20$0.0672315160K$0.68 at Novita AI+102%Sep 30
Qwen3 Omni 30B A3B InstructQwenSep 24—$0.25$0.97—1—64K——Oct 8
Qwen3 Omni 30B A3B ThinkingQwenSep 24—$0.25$0.97—1—64K——Oct 23
Qwen3 MaxQwenSep 23142$1.20$6.00—2—256K$10.56 at Novita AI+47%Oct 23
Qwen3 VL 235B A22B Instruct openAlibabaSep 23—$0.20$0.88$0.10543256K$2.11 at Parasail+95%Oct 23
Qwen3 VL 235B A22B Thinking openAlibabaSep 23—$0.40$4.00—2—128K$4.93 at Novita AI+12%Oct 9
DeepSeek V3.1 Terminus open −40% at StreamLakeDeepSeekSep 22—$0.27$1.00$0.14445160K$1.37 at StreamLake+8%Oct 23
Qwen3 Next 80B A3B Thinking openAlibabaSep 11—$0.15$1.20—2—256K$1.35 at Alibaba+0%Oct 21
Hermes 4 405BNous ResearchAug 26—$1.00$3.00—142128K——Oct 23
DeepSeek V3.1 openDeepSeekAug 21140$0.25$0.95$0.13682160K$2.20 at CoreWeave+83%Oct 23
Baichuan M2 32BBaichuanAug 13—$0.07$0.07—1—128K——Oct 23
Mistral Medium 3.1Mistral AIAug 13—$0.40$2.00$0.04179128K——Oct 23
GLM 4.5V openZ.aiAug 11—$0.60$1.80$0.1125064K$2.40 at Z.AI+0%Oct 23
Claude Opus 4.1AnthropicAug 5144$15.00$75.00$1.501—200K——Oct 17
gpt-oss-120b openOpenAIAug 5140$0.03$0.15$0.01219932128K$1.10 at Cerebras+511%Oct 23
gpt-oss-20b openOpenAIAug 5138$0.02$0.10$0.0213389128K$0.38 at Groq+212%Oct 23
Qwen3 Coder 30B A3B Instruct openAlibabaJul 31—$0.07$0.27—4115256K$2.70 at Alibaba+694%Oct 8
Codestral 2508Mistral AIJul 30—$0.30$0.90$0.031120250K——Oct 23
Qwen3 30B A3B Thinking 2507 openAlibabaJul 30140$0.20$2.40—1—80K——Oct 9
GLM 4.5 openZ.aiJul 28—$0.60$2.20$0.11123128K——Oct 23
GLM 4.5 Air openZ.aiJul 28—$0.13$0.85$0.025361128K$1.30 at Z.AI+33%Oct 23
Qwen3 Coder Flash openQwenJul 28—$0.30$1.50$0.061761M——Oct 23
Qwen3 235B A22B Thinking 2507 openAlibabaJul 25144$0.23$2.30—330128K$3.95 at Venice+56%Oct 9
Qwen3 Coder Plus openQwenJul 23—$1.00$5.00—1—1M——Oct 23
Qwen3 Coder openAlibabaJul 22—$0.30$1.00$0.04590256K$5.85 at Alibaba+350%Oct 23
Qwen3 235B A22B Instruct 2507 open −75% at GMICloudAlibabaJul 21139$0.0875$0.35$0.01751035256K$1.10 at Google+151%Oct 19
Voxtral Small 24B 2507 openMistral AIJul 15—$0.10$0.30$0.0118432K——Oct 23
UncensoredDolphinJul 9—$0.20$0.90—163125K——Oct 23
ERNIE 4.5 21B A3BBaiduJun 30—$0.07$0.28—1—120K——Oct 23
ERNIE 4.5 VL 424B A47BBaiduJun 30—$0.42$1.25—1—123K——Oct 8
Mistral Small 3.2 24B openMistral AIJun 20—$0.075$0.20$0.014130250K$0.40 at Mistral AI+45%Oct 23
Gemini 2.5 FlashGoogleJun 17141$0.30$2.50$0.031—1M——Oct 17
Gemini 2.5 ProGoogleJun 17145$1.25$10.00$0.131—1M——Oct 17
MiniMax M1 openMiniMaxJun 17—$0.40$2.20—2751M$2.75 at Novita AI+6%Oct 23
Mistral MediumMistral AIMay 7—$1.50$7.50—1—128K——Sep 3
Mistral Medium 3Mistral AIMay 7134$0.40$2.00$0.04183128K——Oct 23
Llama Guard 4 12B openMetaApr 30—$0.18$0.18—1—160K——Oct 23
Qwen3 32B openAlibabaApr 30138$0.08$0.28—245128K$0.71 at SiliconFlow+97%Oct 23
Qwen3 8BQwenApr 30136$0.18$0.70—1—128K——Oct 23
Qwen3 14B openAlibabaApr 28138$0.10$0.22—364128K$1.75 at Alibaba+447%Oct 23
Qwen3 30B A3B open −55% at StreamLakeAlibabaApr 28136$0.0481$0.19$0.107154256K$0.65 at Alibaba+170%Oct 9
GPT-4.1OpenAIApr 14137$2.00$8.00$0.501—1.05M——Oct 17
GPT-4.1 miniOpenAIApr 14135$0.40$1.60$0.101831.05M——Oct 23
Llama 4 Maverick openMetaApr 5132$0.20$0.80$0.055681M$1.50 at Google+50%Oct 1
Llama 4 Scout openMetaApr 5130$0.10$0.30—3351.25M$0.95 at Google+137%Oct 23
DeepSeek V3 0324 openDeepSeekMar 24—$0.24$0.90$0.11425160K$2.00 at Crusoe+75%Sep 29
Mistral Small 3.1 24BMistralMar 17128$0.35$0.56—154125K——Oct 23
Command A openCohereMar 13—$2.50$10.00—1—250K——Sep 3
Gemma 3 12B openGoogleMar 12124$0.05$0.10$0.05332128K$0.28 at NextBit+83%Oct 23
Gemma 3 27B openGoogleMar 12130$0.08$0.16$0.04539128K$0.53 at Parasail+121%Oct 23
Gemma 3 4B openGoogleMar 12116$0.04$0.05$0.022123128K$0.15 at DeepInfra+67%Oct 23
Reka Flash 3RekaMar 12—$0.10$0.20—18064K——Oct 23
SabaMistral AIFeb 17—$0.20$0.60$0.0213432K——Oct 23
Qwen2.5 VL 72B InstructQwenFeb 1—$0.80$1.00$0.40137125K——Oct 23
Mistral Small 3MistralJan 30—$0.05$0.08—15532K——Oct 23
DeepSeek R1 openDeepSeekJan 20139$0.50$2.15$0.35443160K$3.20 at Novita AI+21%Oct 23
MiniMax-01MiniMaxJan 15—$0.20$1.10—1251.0M——Oct 23
DeepSeek Chat open −10% at StreamLakeDeepSeekDec 26146$0.32$0.89—228160K$1.29 at StreamLake+6%Oct 23
DeepSeek V3 open −10% at StreamLakeDeepSeekDec 26132$0.32$0.89—228160K$1.29 at StreamLake+6%Oct 23
Phi 4MicrosoftDec 11130$0.07$0.14—17916K——Oct 23
Llama 3.3 70B open −25% at SambaNovaMetaDec 6127$0.10$0.32$0.1012229128K$2.55 at Cloudflare+506%Oct 23
Command R7B (12-2024) openCohereDec 2—$0.0375$0.15—160125K——Oct 23
Qwen3 Next 80B A3B openAlibabaDec 1—$0.0975$0.78$0.075139256K$1.65 at Novita AI+88%Oct 9
Qwen2.5 Coder 32B InstructQwenNov 11119$0.66$1.00—13332K——Oct 23
Qwen2.5 7B InstructQwenOct 16119$0.10$0.20—112832K——Oct 23
Llama 3.2 1B Instruct openMetaSep 25—$0.027$0.20—112560K——Oct 23
Llama 3.2 3B Instruct openMetaSep 25—$0.05$0.33—2207128K$0.39 at Cloudflare+2%Oct 23
Qwen 2.5 72B openAlibabaSep 19129$0.36$0.40—22432K$0.78 at Novita AI+3%Oct 23
Command R openCohereAug 30—$0.15$0.60—1—125K——Sep 3
Command R+ (08-2024) openCohereAug 30119$2.50$10.00—110125K——Oct 23
Hermes 3 70B Instruct openNous ResearchAug 18—$0.70$0.70—131128K——Oct 23
Hermes 3 405B Instruct openNous ResearchAug 16—$1.00$1.00—125128K——Oct 23
Llama 3.1 70B Instruct openMetaJul 23126$0.40$0.40$0.80348128K$1.60 at CoreWeave+100%Oct 23
Llama 3.1 8B openMetaJul 23116$0.02$0.04$0.0255382128K$0.44 at CoreWeave+633%Oct 23
GPT-4o miniOpenAIJul 18126$0.15$0.60$0.075173125K——Oct 23
Gemma 2 27B openGoogleJul 13122$0.65$0.65—1428K——Oct 23
Mistral Nemo open −28% at Io NetMistral AIJul 1119$0.0166$0.0216$0.01086107128K$0.30 at Mistral AI+685%Oct 23
GPT-4oOpenAIMay 13—$2.50$10.00$1.25186125K——Oct 23
Mixtral 8x22B Instruct openMistral AIApr 17122$2.00$6.00$0.20119664K——Oct 23
Qwen Plus 0728QwenJan 25—$0.40$1.20$0.081—1M——Oct 23
GLM-4 32B openZ.aiJan 1—$0.55$1.66—1—32K——Oct 8

Lowest price across providers, per million tokens: input and output of the cheapest offer, the lowest cached-input price among the hosts that publish one · Intel.: Epoch Capabilities Index by Epoch AI (CC BY 4.0), higher is better, no score yet shows — ·tok/s: output speed at the fastest host, OpenRouter, median of the last 30 min, read Oct 23, 06:00 UTC · context, priciest host and the gap between hosts under Columns.

Intelligence for the price

111 models with a score, each at its cheapest host. The line joins the models no other beats on both price and score.

Epoch Capabilities Index

Mistral Nemo Io Netgpt-oss-20b AkashMLQwen3.7 Flash AlibabaGLM-5.3-Flash StreamLakeDeepSeek V4 Flash 0731 RelaceDeepSeek V4.1 Flash Io NetClaude Haiku 5.5 AnthropicGPT-6.1 Sol OpenAIClaude Opus 5.5 Anthropic

Price, input + output per 1M tokens (log) →

On the best-deal line

  1. Claude Opus 5.5 Anthropic
  2. GPT-6.1 Sol OpenAI
  3. Claude Haiku 5.5 Anthropic
  4. DeepSeek V4.1 Flash Io Net
  5. DeepSeek V4 Flash 0731 Relace
  6. GLM-5.3-Flash StreamLake
  7. Qwen3.7 Flash Alibaba
  8. gpt-oss-20b AkashML
  9. Mistral Nemo Io Net

Score: Epoch Capabilities Index, Epoch AI, CC BY 4.0; the axis starts at 100, the scale has no zero · prices read Oct 23, 09:51 UTC.

One line per model. Models on two or more hosts open a provider comparison; single-host models open their offer page. Both carry the maker's declared spec sheet. Intelligence, where shown, is a third-party score we report but do not produce. For the full ranking by gap between hosts, see same model, different price; for quality per dollar, the best value board and the price-of-intelligence chart.

Links can filter this page by the maker's headquarters (?origin=United States, China, Canada, France) or by published size (?size=500B+, 100–500B, under 100B). Maker headquarters come from a first-party source only; size is the maker's published total parameter count.

LLM API pricing FAQ

What is the cheapest LLM API right now?

Mistral Nemo is the lowest-priced model in the index at $0.038 per 1M tokens (input + output) at Io Net, as of 2026-10-23. The cheapest token is not always the cheapest task, though — for quality per dollar, see the value board.

Why does the same model cost different amounts?

Because the thing that carries a price is not the model but the endpoint: the same weights, served by several hosts, at several prices. Across the 112 models served by two or more hosts, the median gap between the cheapest and the priciest host is 62%, and the widest gap is 1347%.

Which price does this table show?

For each model, the input and output price of its cheapest offer (lowest input + output per million tokens), and the lowest cached-input price among the hosts that publish one. The priciest host and the gap between the two are under Columns.

How fresh are these prices?

Figures are as of 2026-10-23. Each price is read from the host's own published page or API and dated; a price that could not be re-verified recently carries a visible warning on its model page instead of a silently aging number.