AI Models & LLMs
Browse and compare AI models from leading organizations. Find the right model for your use case based on pricing, capabilities, and context window.5 models listed
Intelligence vs. Cost
Average benchmark score plotted against cost per million tokens. Dot size reflects context window.
OrganizationAllAI21 LabsAI4Finance FoundationAlibabaAllenAIAmazonAnthropicApodexArcee AIBaiduBlack Forest LabsByteDanceCactus ComputeCohereCompVisDeepGroveDeepSeekGoogleIBMInceptionIndexTeamInflection AIInstituto IA/LNCCJingyao GongK2-FSAKyutaiLightricksLiquid AIMetaMicrosoftMiniMaxMistral AIMoonshot AINVIDIANous ResearchOpenAIOpenBMBOpenMOSSPerplexity AIPrior LabsRekaRobloxRoboflowStepFunSunoTencentTsinghua UniversityUpstageWriterXiaomiZ.aixAI
Capability3d-generationadaptive-thinkingagent-decisioningagenticagentic-codingagentic-workflowsagentsanomaly-detectionaudioaudio-generationaudio-inputaudio-understandingbrowsingcinematic-controlclassificationcodecodingcomputer-useconfidencecontext-cachingdevice-controledge-inferenceeducationembeddingsemotion-controlextended-thinkingfile-searchfinance-nlpfinancial-modelingfine-tuningformal-verificationfunction-callingimage-generationimage-segmentationimage-text-retrievalimage-to-imageimage-to-videoimage-understandinginstance-segmentationjson-modejson-schemakeypoint-detectionlocal-deploymentlocal-inferencelong-contextlora-trainingmulti-reference editingmultilingualmultilingual-audiomultilingual-speechmultimodalmultimodal-inputobject-detectionofflineon-deviceon-device-inferenceopen-weightspart-controllable-generationpdf-inputportugueseprobability-scoringpromptable-segmentationpronunciation-controlquantile-forecastingreasoningregressionresearchresponses-apiroutingscoringsearch-groundingself-supervised-learningsentiment-analysisshape-generationsound-effectsspeech-generationspeed-controlstreamingstreaming-inferencestreaming-ttsstructured-outputsynthetic-datatexttext-generationtext-to-3dtext-to-audiotext-to-imagetext-to-speechtext-to-videotime-series-forecastingtool-usetrainingvideovideo-editingvideo-extensionvideo-generationvideo-inputvideo-segmentationvideo-understandingvisionvoice-cloningvoice-cloning-style-promptsvoice-designvolatility-predictionweb-searchzero-shot-classificationzero-shot-forecasting
Qwen3.6 PlusMultimodal LLM
qwen/qwen3.6-plus
89
78.2
89
78.2
Qwen3.5-9BMultimodal LLM
qwen/qwen3.5-9b
26
204
26
204
Qwen3.5-35B-A3BMultimodal LLM
qwen/qwen3.5-35b-a3b
39
52.8
39
52.8
Qwen3.5-27BMultimodal LLM
qwen/qwen3.5-27b
57
64.5
57
64.5
Qwen3-VLMultimodal LLM
qwen-vl-max
$0.20/M / $0.60/M131.1K ctx
$0.20/M / $0.60/M131.1K ctx