LLM AtlasLLM AtlasSearch models

Leaderboard

enterprise rankings

Updated weekly

Each leaderboard uses transparent weighted scoring, current model context, and supporting analysis to help teams interpret the results with confidence. Only full-profile entries appear in rankings; broader catalog records remain available elsewhere on the site when only source-backed metadata is currently available.

Full-profile entries
245

Models with complete enough metadata and scoring coverage to be meaningfully ranked in this category.

Ranking basis
enterprise

Scores combine benchmark evidence, product metadata, and cost/context signals when those fields are published.

Catalog caveat
25

Tracked models without full scoring remain in the directory and provider pages, but are not relied on for analytical ranking claims.

RankModelProviderScoreContext
#1Claude 3.7 SonnetAnthropic92200,000
#2Claude Sonnet 4.5Anthropic921,000,000
#3Claude Sonnet 4.6Anthropic921,000,000
#4Claude Opus 4.6Anthropic911,000,000
#5Claude Sonnet 4Anthropic901,000,000
#6GPT-4oOpenAI90128,000
#7GPT-5.4OpenAI901,000,000
#8Command R+ 2026Cohere89128,000
#9Gemini 2.5 ProGoogle DeepMind891,048,576
#10Gemini 3.1 FlashGoogle DeepMind891,048,576
#11Gemini 3.1 ProGoogle DeepMind891,048,576
#12GPT-5.4 ProOpenAI891,000,000
#13Claude Haiku 4.5Anthropic88200,000
#14Claude Opus 4Anthropic88200,000
#15Claude Opus 4.1Anthropic88200,000
#16Gemini 3.0 ProGoogle DeepMind881,048,576
#17GPT-5.2OpenAI881,000,000
#18GPT-5.3 InstantOpenAI881,000,000
#19Gemini 2.5 FlashGoogle DeepMind871,048,576
#20Gemini 2.5 Pro TTSGoogle DeepMind871,048,576
#21Gemini 3.0 FlashGoogle DeepMind871,048,576
#22GPT-5OpenAI871,000,000
#23GPT-5 miniOpenAI871,000,000
#24GPT-5.2 ProOpenAI871,000,000
#25GPT-5.3-CodexOpenAI871,000,000
#26Gemini 2.0 FlashGoogle DeepMind861,048,576
#27Gemini 3.1 Flash-LiteGoogle DeepMind861,048,576
#28Mistral Large 25Mistral AI86128,000
#29Gemini 2.5 Flash LiveGoogle DeepMind851,048,576
#30Gemini 2.5 Flash Native Audio PreviewGoogle DeepMind851,048,576
#31Gemini 2.5 Flash-LiteGoogle DeepMind851,048,576
#32GPT-4.1OpenAI851,048,576
#33GPT-4o-miniOpenAI85128,000
#34Sonar Deep ResearchPerplexity85200,000
#35Sonar Reasoning ProPerplexity85200,000
#36GPT-4.1 miniOpenAI841,048,576
#37GPT-5 nanoOpenAI841,000,000
#38Nova ProAmazon Web Services84300,000
#39o4-miniOpenAI84200,000
#40Sonar ProPerplexity84200,000
#41Claude Haiku 3.5Anthropic83200,000
#42Claude Sonnet 3Anthropic83200,000
#43Gemini 1.5 ProGoogle DeepMind832,097,152
#44Gemini 2.0 Flash-LiteGoogle DeepMind831,048,576
#45Llama 4 MaverickMeta831,048,576
#46SonarPerplexity83128,000
#47warpgrep-v2Morph83128,000
#48Claude Opus 3Anthropic82200,000
#49Gemini 1.5 FlashGoogle DeepMind821,048,576
#50morph-v3-fast-applyMorph82128,000
#51Nova LiteAmazon Web Services82300,000
#52o4-mini-deep-researchOpenAI82200,000
#53Nova MicroAmazon Web Services81128,000
#54Claude Haiku 3Anthropic80200,000
#55flash-compactMorph80200,000
#56gpt-audioOpenAI80128,000
#57gpt-realtimeOpenAI80128,000
#58o1-miniOpenAI80128,000
#59o3OpenAI80200,000
#60o3-deep-researchOpenAI80200,000
#61Gemini 1.5 Flash-8BGoogle DeepMind791,048,576
#62o1OpenAI77200,000
#63Claude Haiku 3Anthropic76200,000
#64Claude Haiku 3.5Anthropic76200,000
#65Claude Haiku 4.5Anthropic76200,000
#66CodestralMistral AI76256,000
#67CogVideoXZ.AI76128,000
#68CogView 4Z.AI76128,000
#69Devstral Medium 1.0Mistral AI76128,000
#70Doubao-Seed-1.6ByteDance / Doubao76128,000
#71Doubao-Seed-1.6-FlashByteDance / Doubao76128,000
#72Doubao-Seed-2.0-CodeByteDance / Doubao76128,000
#73Doubao-Seed-CodeByteDance / Doubao76128,000
#74ERNIE 3.5 128KBaidu / ERNIE76128,000
#75ERNIE 4.0 Turbo 8KBaidu / ERNIE76128,000
#76ERNIE Functions 8KBaidu / ERNIE76128,000
#77ERNIE Speed 128KBaidu / ERNIE76128,000
#78GLM-4.5Z.AI76128,000
#79GLM-4.5VZ.AI76128,000
#80GLM-4.6Z.AI76128,000
#81GLM-4.6VZ.AI76128,000
#82GLM-4.7Z.AI76128,000
#83GLM-5Z.AI76128,000
#84GLM-ImageZ.AI76128,000
#85GLM-OCRZ.AI76128,000
#86gpt-audio-miniOpenAI76128,000
#87gpt-realtime-miniOpenAI76128,000
#88Hunyuan CodeTencent / Hunyuan76128,000
#89Hunyuan LiteTencent / Hunyuan76128,000
#90Hunyuan StandardTencent / Hunyuan76128,000
#91Hunyuan T1Tencent / Hunyuan76256,000
#92Hunyuan T1 VisionTencent / Hunyuan76128,000
#93Hunyuan TurboSTencent / Hunyuan76128,000
#94Hunyuan TurboS LongText 128KTencent / Hunyuan76128,000
#95Kimi K2Moonshot AI / Kimi76131,072
#96Kimi K2 ThinkingMoonshot AI / Kimi76256,000
#97Kimi K2 Turbo PreviewMoonshot AI / Kimi76256,000
#98Kimi K2.5Moonshot AI / Kimi76256,000
#99Magistral Medium 1.2Mistral AI76128,000
#100MiniMax-M1MiniMax76204,800
#101MiniMax-Text-01MiniMax76204,800
#102MiniMax-VL-01MiniMax76204,800
#103Mistral Large 3Mistral AI76128,000
#104Mistral Medium 3.1Mistral AI76128,000
#105Mistral Small 3.1Mistral AI76128,000
#106Mistral Small 3.2 OpenMistral AI76128,000
#107Pixtral 12BMistral AI76131,072
#108Pixtral LargeMistral AI76131,072
#109Vidu Q1Z.AI76128,000
#110Voxtral Mini OpenMistral AI76131,072
#111Voxtral Mini TranscribeMistral AI76131,072
#112Voxtral Small OpenMistral AI76131,072
#113Claude Sonnet 3Anthropic75200,000
#114Claude Sonnet 4Anthropic751,000,000
#115Codestral EmbedMistral AI7532,768
#116Command ACohere75256,000
#117Command A ReasoningCohere75128,000
#118Command A TranslateCohere75128,000
#119Command A VisionCohere75128,000
#120Command R+Cohere75128,000
#121Command R7BCohere75128,000
#122DeepSeek-Coder-V2DeepSeek75128,000
#123DeepSeek-Math-V2DeepSeek75128,000
#124DeepSeek-R1DeepSeek75128,000
#125DeepSeek-R1-Distill-Llama-70BDeepSeek75128,000
#126DeepSeek-V2.5DeepSeek75128,000
#127DeepSeek-V3DeepSeek75128,000
#128DeepSeek-V3.1DeepSeek75128,000
#129DeepSeek-V3.1-BaseDeepSeek75128,000
#130DeepSeek-V3.2DeepSeek75128,000
#131DeepSeek-V3.2-ExpDeepSeek75128,000
#132Devstral 2 OpenMistral AI75128,000
#133Devstral Small 2Mistral AI75128,000
#134Embed 4Cohere75128,000
#135ERNIE 4.5 Turbo 32KBaidu / ERNIE7532,768
#136GPT Image 1OpenAI7532,768
#137Grok 3xAI75131,072
#138Grok 3 MinixAI75131,072
#139Grok 4xAI75256,000
#140Grok 4 Fast ReasoningxAI75131,072
#141grok-imagexAI75131,072
#142Jamba 3BAI21 Labs75256,000
#143Jamba LargeAI21 Labs75256,000
#144Jamba Large 1.6AI21 Labs75256,000
#145Jamba MiniAI21 Labs75256,000
#146Jamba Mini 1.6AI21 Labs75256,000
#147Jamba Mini 1.7AI21 Labs75256,000
#148Llama Guard 4 12BMeta75131,072
#149Magistral Small 1.2 OpenMistral AI75128,000
#150MiniMax-M2MiniMax75204,800
#151MiniMax-M2.1MiniMax75204,800
#152MiniMax-M2.1-highspeedMiniMax75204,800
#153MiniMax-M2.5MiniMax75204,800
#154MiniMax-M2.5-highspeedMiniMax75204,800
#155Ministral 3 14B OpenMistral AI75128,000
#156Ministral 3 3B OpenMistral AI75128,000
#157Ministral 3 8B OpenMistral AI75128,000
#158Mistral EmbedMistral AI7532,768
#159Mistral Large 3 OpenMistral AI75128,000
#160Mistral ModerationMistral AI7532,768
#161Mistral Nemo 12BMistral AI75128,000
#162Mistral OCR 2505Mistral AI7532,768
#163Phi-3-vision-128k-instructMicrosoft75128,000
#164Phi-3.5-mini-instructMicrosoft75131,072
#165Phi-3.5-MoE-instructMicrosoft75131,072
#166Phi-3.5-vision-instructMicrosoft75131,072
#167Phi-4-mini-flash-reasoningMicrosoft75131,072
#168Phi-4-mini-instructMicrosoft75131,072
#169Phi-4-multimodal-instructMicrosoft75131,072
#170Phi-4-reasoningMicrosoft75131,072
#171Phi-4-reasoning-plusMicrosoft75131,072
#172Phi-4-reasoning-vision-15BMicrosoft75131,072
#173Qwen2.5-1.5B-InstructAlibaba Qwen75131,072
#174Qwen2.5-14B-InstructAlibaba Qwen75131,072
#175Qwen2.5-32B-InstructAlibaba Qwen75131,072
#176Qwen2.5-3B-InstructAlibaba Qwen75131,072
#177Qwen2.5-72B-InstructAlibaba Qwen75131,072
#178Qwen2.5-7B-InstructAlibaba Qwen75131,072
#179Qwen2.5-MaxAlibaba Qwen75131,072
#180Qwen2.5-VL-72B-InstructAlibaba Qwen75131,072
#181Qwen2.5-VL-7B-InstructAlibaba Qwen75131,072
#182Qwen3-Coder-NextAlibaba Qwen75131,072
#183Qwen3.5-0.8BAlibaba Qwen75131,072
#184Qwen3.5-122B-A10BAlibaba Qwen75131,072
#185Qwen3.5-27BAlibaba Qwen75131,072
#186Qwen3.5-2BAlibaba Qwen75131,072
#187Qwen3.5-35B-A3BAlibaba Qwen75131,072
#188Qwen3.5-397B-A17BAlibaba Qwen75131,072
#189Qwen3.5-4BAlibaba Qwen75131,072
#190Qwen3.5-9BAlibaba Qwen75131,072
#191chatgpt-image-latestOpenAI7432,768
#192GPT-4o mini TranscribeOpenAI74128,000
#193GPT-4o mini TTSOpenAI74128,000
#194GPT-4o TranscribeOpenAI74128,000
#195gpt-image-1-miniOpenAI7432,768
#196image-01MiniMax748,192
#197image-01-liveMiniMax748,192
#198Llama 3.3 70B InstructMeta74128,000
#199MiniMax-Speech-02MiniMax748,192
#200music-2.0MiniMax748,192
#201DeepSeek-OCRDeepSeek7316,384
#202DeepSeek-OCR-2DeepSeek7316,384
#203DeepSeek-VL2-SmallDeepSeek7316,384
#204gpt-oss-120bOpenAI73131,072
#205Janus-Pro-7BDeepSeek7316,384
#206Phi-4Microsoft7316,384
#207phi-1Microsoft724,096
#208phi-1_5Microsoft724,096
#209phi-2Microsoft724,096
#210Phi-3-medium-4k-instructMicrosoft724,096
#211Phi-3-mini-4k-instructMicrosoft724,096
#212Phi-tiny-MoE-instructMicrosoft724,096
#213Claude Opus 3Anthropic71200,000
#214Claude Opus 4Anthropic71200,000
#215LFM2-24B-A2BLiquid AI7132,768
#216Llama 3.1 405B InstructMeta71128,000
#217Llama 3.1 70B InstructMeta71128,000
#218Llama Guard 3 11B VisionMeta71131,072
#219pplx-embed-v1-4bPerplexity718,192
#220Step-3.5-FlashStepFun71131,072
#221Llama 4 ScoutMeta7010,485,760
#222pplx-embed-v1-0.6bPerplexity708,192
#223gpt-oss-20bOpenAI68131,072
#224LFM2-8B-A1BLiquid AI6732,768
#225Llama 3.1 8B InstructMeta67128,000
#226Llama 3.2 90B Vision InstructMeta67128,000
#227MiMo-VL-7BXiaomi67131,072
#228Step3-VL-10BStepFun67131,072
#229Llama 3.2 1B InstructMeta65128,000
#230Llama 3.2 3B InstructMeta65128,000
#231MiMo-Audio-7BXiaomi65131,072
#232Code Llama 70B InstructMeta648,192
#233Llama 3.2 11B Vision InstructMeta64128,000
#234Meta Llama 3 70B InstructMeta648,192
#235Step-Audio-R1.1StepFun64131,072
#236FLUX 1.1 ProBlack Forest Labs63512
#237LFM2.5-1.2B-InstructLiquid AI63131,072
#238FLUX 1.1 Pro UltraBlack Forest Labs62512
#239LFM2.5-1.2B-ThinkingLiquid AI62131,072
#240Prompt Guard 86MMeta62512
#241Code Llama 34B InstructMeta618,192
#242FLUX 1 ProBlack Forest Labs61512
#243Meta Llama 3 8B InstructMeta618,192
#244LFM2-2.6BLiquid AI5932,768
#245NextStep-1.1StepFun50512

Why #1: Claude 3.7 Sonnet

A top-tier reasoning model with strong software engineering assistance and enterprise controls.

This model clears the current full-profile threshold for leaderboard methodology.

Why #2: Claude Sonnet 4.5

Anthropic's Sonnet 4.5 with 1M token context for fast frontier reasoning, coding, and long-context agent work.

This model clears the current full-profile threshold for leaderboard methodology.

Why #3: Claude Sonnet 4.6

Anthropic's current Sonnet tier for fast frontier reasoning, coding, and long-context agent work.

This model clears the current full-profile threshold for leaderboard methodology.