Text & knowledge

Ask, research, write, translate, read documents and search your own. 6 tasks, 457 listed options.

Tier: how close to the best result on the task's main public benchmark
Outlined: a tool inherits the tier of the best model it offers
No public result yet
Cloud · Local: runs on your own hardware

Chat & assistants

Ask, write, research and reason with a general AI assistant.

Best overallSGemini 4 Argon
Best valueAMiMo-V2.6-Flash
Best free startSGemini
Best local (≤24 GB GPU)AGemma 4 (31B)

Models

Local

Kimi K3Moonshot AI · ~1600 GB VRAM$A

MiMo-V2.6-ProXiaomi · ~585 GB VRAM$A

GLM-5.3Z.ai · ~425 GB VRAM$A

DeepSeek-V4.1-Flash-1017DeepSeek · ~320 GB VRAM$A

GLM-5.3-FlashZ.ai · ~185 GB VRAM$A

DeepSeek-V4-Pro-0813DeepSeek · ~920 GB VRAM A

Gemma 4 (31B)Google DeepMind · ~19 GB VRAM$A

MiMo-V2.6-FlashXiaomi · ~178 GB VRAM A

Step 5 PreviewStepFun · ~346 GB VRAM A

MiniMax-M3MiniMax · ~245 GB VRAM A

InklingThinking Machines Lab · ~557 GB VRAM A

Qwen3.8-27BAlibaba (Qwen) · ~17 GB VRAM$A

Mistral Medium 3.5Mistral AI · ~75 GB VRAM A

Nemotron 3 Ultra 550B-A55BNVIDIA · ~321 GB VRAM A

Muse Glimmer 30BMeta · ~19 GB VRAM B

Nemotron 3.5 Super VL 120B-A12BNVIDIA · ~72 GB VRAM B

Inkling SmallThinking Machines Lab · ~159 GB VRAM B

Trinity Large ThinkingArcee AI · ~229 GB VRAM B

Ling-3.1-flashinclusionAI (Ant Group) · ~321 GB VRAM B

gpt-oss-120bOpenAI · ~80 GB VRAM B

Nemotron 3.5 Lightning 30B-A3BNVIDIA · ~20 GB VRAM B

Qwen3.8-2.4T-A95BAlibaba (Qwen) · ~1370 GB VRAM B

Qwen3.8-Flash-NextAlibaba (Qwen) · ~104 GB VRAM C

Granite 4.2 30BIBM · ~18 GB VRAM C

Muse Spark Open (816B-A42B)Meta · ~470 GB VRAM C

Olmo 3.1 32BAllen Institute for AI (Ai2) · ~20 GB VRAM C

Llama 4 MaverickMeta · ~230 GB VRAM C

gpt-oss-20bOpenAI · ~16 GB VRAM$C

Motif-3Motif Technologies · ~181 GB VRAM C

K2-Horizon-375B-A23BMBZUAI Institute of Foundation Models · ~218 GB VRAM C

Beam 1.0 (501B-A23B)Reflection AI · ~287 GB VRAM C

Kimi K2.7 CodeMoonshot AI · ~587 GB VRAM C

K2-Horizon-MoVA-36B-A4BMBZUAI Institute of Foundation Models · ~23 GB VRAM C

Solar Open2 250BUpstage · ~144 GB VRAM C

A.X K2SK Telecom · ~396 GB VRAM C

Nemotron 3.5 Bolt 74B-A9BNVIDIA · ~24 GB VRAM C

K-EXAONE 2.0 (750B-A37B)LG AI Research · ~428 GB VRAM C

LongCat-2.0Meituan (LongCat) · ~1014 GB VRAM C

Command A+ (05-2026)Cohere · ~126 GB VRAM C

Ministral 3 14B (2512)Mistral AI · ~9 GB VRAM$C

Mistral Small 4 (2603)Mistral AI · ~69 GB VRAM C

Ring-2.6-1TinclusionAI (Ant Group) · ~586 GB VRAM C

Sarvam-105BSarvam AI · ~62 GB VRAM –

Hy4-previewTencent · ~441 GB VRAM –

Laguna M.1Poolside · ~130 GB VRAM –

LFM2.5-24B-A2BLiquid AI · ~15 GB VRAM

Apps

Cloud

ChatLLMAbacus.AI · $10/mo S

GeminiGoogle · Free tier · $4.99/mo$S

MammouthMammouth AI S

PerplexityPerplexity AI · Free tier · $20/mo S

VeniceVenice.ai · Free tier · $18/mo S

PoeQuora · Free tier S ChatWiseChatWise S

ClaudeAnthropic · Free tier · $20/mo S

Google AI StudioGoogle · Free$S

Meta AIMeta · Free tier S HuggingChatHugging Face · Free tier · $9/mo S

Kagi AssistantKagi · Free tier · $5/mo S

KimiMoonshot AI · Free tier S

ChatGPTOpenAI · Free tier · $20/mo S Microsoft CopilotMicrosoft · Free tier · $9.99/mo A

Z.ai ChatZ.ai A Wenxin Assistant (ERNIE Bot)Baidu A

DeepSeekDeepSeek · Free A

YuanbaoTencent · Free A

Duck.aiDuckDuckGo · Free tier A

GrokSpaceXSI (formerly xAI) · Free tier · $8/mo A

Vibe (formerly Le Chat)Mistral AI · Free tier · $15/mo B

Qwen ChatAlibaba (Qwen team) · Free –

Alexa+Amazon · Free tier · $20/mo –

DoubaoByteDance · Free tier –

iFlytek Spark (Xinghuo)iFlytek –

LumoProton · Free tier · $13/mo –

Local

Open WebUIOpen WebUI Inc. · Free$S

NextChatNextChat · Free$S

LobeHub (formerly LobeChat)LobeHub · Free tier · $9.9/mo S

AnythingLLMMintplex Labs · Free$S

Cherry StudioCherryHQ · Free$S

TextGen (text-generation-webui)oobabooga · Free$S

LibreChatLibreChat · Free$S

JanMenlo Research · Free$S

ChatboxChatbox AI · Free tier · $3.99/mo S

KhojKhoj · Free$S

Big-AGIBig-AGI · Free$S

DeepChatThinkInAI · Free$S

5irenanbingxyz · Free$S

WitsyKochava Studios · Free$S

LM StudioElement Labs · Free$S

Msty StudioMsty · Free$A

Unsloth (Studio / Desktop app)Unsloth AI · Free A

Page Assistn4ze3m · Free A

OsaurusOsaurus · Free A

Enchantedgluonfield · Free A AlpacaJeffser · Free A

TypingMindTypingMind A

Google AI Edge GalleryGoogle · Free A

PocketPal AIa-ghorbani · Free A

ChatterUIVali-98 · Free A

MaidMobile Artificial Intelligence · Free A

Llamaggml-org and Hugging Face · Free A

Locally AIElement Labs (LM Studio)–

Brave LeoBrave · Free tier · $15/mo

Runtimes & gateways

Cloud

Gemini Enterprise Agent Platform (formerly Vertex AI)Google Cloud S

OpenRouterOpenRouter · Free tier S

PerplexityPerplexity AI · Free tier · $20/mo$S

RequestyRequesty S

Vercel AI GatewayVercel · Free tier S

Amazon BedrockAmazon Web Services S

DeepInfraDeepInfra S

Microsoft FoundryMicrosoft S

Fireworks AIFireworks AI S

Hugging Face Inference ProvidersHugging Face · Free tier · $9/mo$S

Novita AINovita AI S

Together AITogether AI A

Cloudflare Workers AICloudflare · Free tier A

NVIDIA NIM APIs (build.nvidia.com)NVIDIA · Free tier A

SambaNova CloudSambaNova A

Cerebras InferenceCerebras · Free tier A

Local

OllamaOllama · Free$S

llama.cppggml-org · Free$S

vLLMvLLM project · Free$S

LiteLLMBerriAI · Free$S

LocalAIEttore Di Giacinto (mudler) · Free$S

exoEXO Labs · Free$S

SGLangSGLang project (LMSYS) · Free$S KTransformersKVCache.AI (Tsinghua MADSys) · Free$S

TensorRT LLMNVIDIA · Free$S

Portkey AI GatewayPortkey · Free$S

XinferenceXorbits · Free$S

LMDeployInternLM (Shanghai AI Laboratory) · Free$S

MLX LMApple (ml-explore) · Free$S

GPUStackGPUStack · Free$A

llamafileMozilla.ai · Free A

LLM (CLI)Simon Willison · Free A

AIChatsigoden · Free A

mistral.rsEric Buehler · Free A

llama-swapmostlygeek · Free A

LemonadeAMD (lemonade-sdk) · Free A

RamaLamaRamaLama project (containers) · Free A

TabbyAPI (ExLlamaV3)theroyallab · Free A

Docker Model RunnerDocker · Free A

MLC LLMMLC AI · Free A

WebLLMMLC AI · Free A GenieXQualcomm · Free C

Foundry LocalMicrosoft · Free

Full chat & assistants map, chart and table →

Search & research

Get answers from the live web with sources, or a full research report.

Best overallSGPT-6.1 Sol
Best valueAGPT-6 Luna
Best free startSExa
Best local (≤24 GB GPU)No qualifying option yet

Models

Cloud

GPT-6.1 SolOpenAI · $4/M tokens$S

GPT-6 AstraOpenAI · $20/M tokens S

Claude Fable 5.1Anthropic · $20/M tokens S Qwen3.8-MaxAlibaba (Qwen) · $3/M tokens$A

Claude Opus 5.5Anthropic · $8/M tokens A

GPT-6 LunaOpenAI · $0.2/M tokens$A

Gemini 3.1 Pro (Preview)Google · $4.5/M tokens A

ERNIE 5.1Baidu B

Sonar ProPerplexity AI · $6/M tokens B

Sonar Reasoning ProPerplexity AI · $3.5/M tokens –

Claude Sonnet 5.5Anthropic · $4/M tokens –

Gemini 3.8 FlashGoogle · $1.5/M tokens –

Grok 4.7xAI · $3/M tokens –

SonarPerplexity AI · $1/M tokens –

Sonar Deep ResearchPerplexity AI · $3.5/M tokens

Local

Nothing tracked yet

Apps

Cloud

SciraScira · Free tier · $15/mo S Microsoft CopilotMicrosoft · Free tier · $9.99/mo$S

ClaudeAnthropic · Free tier · $20/mo A

GeminiGoogle · Free tier · $4.99/mo$A

ChatGPTOpenAI · Free tier · $20/mo A

Kagi AssistantKagi · Free tier · $5/mo$B

PerplexityPerplexity AI · Free tier · $20/mo B

GrokSpaceXSI (formerly xAI) · Free tier · $8/mo –

Brave SearchBrave –

ConsensusConsensus –

DeepSeekDeepSeek · Free –

DoubaoByteDance · Free tier –

ElicitElicit · Free tier · $49/mo –

FeloFelo · Free tier · $15/mo –

Google AI ModeGoogle · Free –

iAskiAsk.Ai · Free tier –

KimiMoonshot AI · Free tier –

KomoKomo · Free tier –

LinerLiner · Free tier · $18/mo –

Meta AIMeta · Free tier –

Metaso AI SearchMetaso (秘塔科技)–

Qwen ChatAlibaba (Qwen team) · Free –

UndermindUndermind · Free tier · $16/mo –

Vibe (formerly Le Chat)Mistral AI · Free tier · $15/mo

Local

Open WebUIOpen WebUI Inc. · Free –

KhojKhoj · Free –

Vane (formerly Perplexica)ItzCrazyKns (open-source project) · Free –

Local Deep ResearchLearningCircuit (open-source project) · Free –

MorphicMorphic (miurla) · Free

Search APIs & engines

Cloud

ExaExa · Free tier S

Perplexity Search APIPerplexity AI S

OpenRouterOpenRouter · Free tier A

ParallelParallel Web Systems · Free tier A

Brave Search APIBrave · Free tier A

You.com APIsYou.com · Free tier B

LinkupLinkup · Free tier B

PerplexityPerplexity AI · Free tier · $20/mo$B

TavilyTavily · Free tier · $30/mo –

Jina Search Foundation APIsJina AI (Elastic) · Free tier –

Local

FirecrawlFirecrawl · Free tier · $19/mo – SearXNGSearXNG (open-source project) · Free –

GPT ResearcherGPT Researcher (open-source project) · Free

Full search & research map, chart and table →

Writing

Draft, rewrite and polish text, from emails to novels.

Best overallSClaude Opus 5.5
Best valueAGLM-5.3-Flash
Best free startSClaude
Best local (≤24 GB GPU)BGemma 4 (31B)

Models

Local

Kimi K3Moonshot AI · ~1600 GB VRAM$A

GLM-5.3Z.ai · ~425 GB VRAM$A

DeepSeek-V4-Pro-0813DeepSeek · ~920 GB VRAM A

MiMo-V2.6-ProXiaomi · ~585 GB VRAM A

DeepSeek-V4.1-Flash-1017DeepSeek · ~320 GB VRAM$A

GLM-5.3-FlashZ.ai · ~185 GB VRAM$B

Gemma 4 (31B)Google DeepMind · ~19 GB VRAM$B

MiniMax-M3MiniMax · ~245 GB VRAM B

Step 5 PreviewStepFun · ~346 GB VRAM B

Mistral Medium 3.5Mistral AI · ~75 GB VRAM B

MiMo-V2.6-FlashXiaomi · ~178 GB VRAM B

InklingThinking Machines Lab · ~557 GB VRAM B

Qwen3.8-27BAlibaba (Qwen) · ~17 GB VRAM$B

Muse Glimmer 30BMeta · ~19 GB VRAM C

Inkling SmallThinking Machines Lab · ~159 GB VRAM C

Llama 4 MaverickMeta · ~230 GB VRAM C

gpt-oss-120bOpenAI · ~80 GB VRAM C

Hemmingway-1The Commons Lab (Altworld) · ~17 GB VRAM C

gpt-oss-20bOpenAI · ~16 GB VRAM$C

Qwen3.8-2.4T-A95BAlibaba (Qwen) · ~1370 GB VRAM

Apps

Cloud

ClaudeAnthropic · Free tier · $20/mo$S

GeminiGoogle · Free tier · $4.99/mo$A

KimiMoonshot AI · Free tier A

Meta AIMeta · Free tier A

ChatGPTOpenAI · Free tier · $20/mo A Microsoft CopilotMicrosoft · Free tier · $9.99/mo A

DeepSeekDeepSeek · Free$A

GrokSpaceXSI (formerly xAI) · Free tier · $8/mo B

Vibe (formerly Le Chat)Mistral AI · Free tier · $15/mo C

HemmingwayThe Commons Lab (Altworld)C

Qwen ChatAlibaba (Qwen team) · Free –

AI DungeonLatitude –

DeepLDeepL · Free tier –

Gemini in Google WorkspaceGoogle –

GrammarlySuperhuman (formerly Grammarly Inc.) · Free tier –

Hemingway EditorHemingway · Free tier –

JasperJasper AI · $69/mo –

Microsoft 365 CopilotMicrosoft · $25/mo –

NotionNotion Labs · Free tier · $24/mo –

NovelAIAnlatan · $10/mo –

NovelcrafterNovelcrafter · $8/mo – ProWritingAidProWritingAid · Free tier – SquiblerSquibler · Free tier · $30/mo –

SudowriteSudowrite · $10/mo –

TypeType.ai · Free tier · $8/mo –

WordtuneAI21 Labs · Free tier · $4.89/mo

Local

Open WebUIOpen WebUI Inc. · Free$A

TextGen (text-generation-webui)oobabooga · Free$A

JanMenlo Research · Free$A

SillyTavernSillyTavern (open-source project) · Free$A

KoboldCppLostRuins (open-source project) · Free$A

LM StudioElement Labs · Free$A

Msty StudioMsty · Free$–

LanguageToolLanguageTooler GmbH · Free tier –

mikupadlmg-anon · Free

Runtimes & gateways

Cloud

OpenRouterOpenRouter · Free tier –

DeepLDeepL · Free tier

Local

OllamaOllama · Free$A

llama.cppggml-org · Free$A

KoboldCppLostRuins (open-source project) · Free$

Full writing map, chart and table →

Translation

Translate text and documents between languages.

No public leaderboard covers this task yet, so its options are mapped but not ranked.

Models

Cloud

qwen-mt-plusAlibaba (Qwen)

Local

Hy-MT2-30B-A3BTencent · ~19 GB VRAM –

North Small Translate 1.0Cohere · ~126 GB VRAM –

TranslateGemma 27BGoogle · ~18 GB VRAM

Apps

Local

– Firefox TranslationsMozilla · Free

APIs & offline engines

Cloud

Amazon TranslateAmazon Web Services · Free tier –

DeepLDeepL · Free tier –

Google TranslateGoogle · Free –

Lara TranslateTranslated · Free tier –

Microsoft TranslatorMicrosoft · Free –

OpenRouterOpenRouter · Free tier – SYSTRAN TranslateSYSTRAN · Free tier

Local

llama.cppggml-org · Free –

LibreTranslateLibreTranslate (open-source project) · Free – Argos TranslateArgos Open Technologies · Free –

CTranslate2OpenNMT · Free

Full translation map, chart and table →

Documents & OCR

Turn PDFs and scans into clean text, tables and data.

Best overallSTeleOCR
Best valueNo qualifying option yet
Best free startNo qualifying option yet
Best local (≤24 GB GPU)STeleOCR

Models

Cloud

Claude Fable 5.1Anthropic · $20/M tokens –

Claude Haiku 5.5Anthropic · $0.2/M tokens –

Claude Opus 5.5Anthropic · $8/M tokens –

Claude Sonnet 5.5Anthropic · $4/M tokens –

Gemini 3.1 Pro (Preview)Google · $4.5/M tokens –

Gemini 3.8 FlashGoogle · $1.5/M tokens –

GPT-6 AstraOpenAI · $20/M tokens –

GPT-6 LunaOpenAI · $0.2/M tokens –

GPT-6.1 SolOpenAI · $4/M tokens –

Mistral OCR 4.1Mistral AI

Local

TeleOCRXingChen-AGI · ~5 GB VRAM$S

OvisOCR2ATH-MaaS · ~4 GB VRAM$S

PaddleOCR-VL-1.6Baidu (PaddlePaddle) · ~4 GB VRAM S

MinerU2.5-Pro-2605-1.2BOpenDataLab (Shanghai AI Laboratory) · ~4 GB VRAM S

GLM-OCRZ.ai · ~5 GB VRAM S

Unlimited-OCRBaidu · ~8 GB VRAM S

Qianfan-OCRBaidu · ~11 GB VRAM S

Youtu-ParsingTencent (Youtu) · ~7 GB VRAM S

FireRed-OCRFireRed Team · ~6 GB VRAM A

DeepSeek-OCR 2DeepSeek · ~9 GB VRAM A

Dolphin-v2ByteDance · ~9 GB VRAM –

Chandra OCR 2Datalab · ~13 GB VRAM –

dots.mocrrednote hi lab (dots studio) · ~8 GB VRAM –

Falcon-OCRTechnology Innovation Institute (TII) · ~2 GB VRAM –

Gemma 4 (31B)Google DeepMind · ~19 GB VRAM –

GLM-5.3-FlashZ.ai · ~185 GB VRAM –

Granite Vision 4.1 4BIBM · ~10 GB VRAM –

granite-docling-258MIBM · ~3 GB VRAM –

HPD-ParsingBaidu (PaddlePaddle) · ~4 GB VRAM –

HunyuanOCR-1.5Tencent (Hunyuan) · ~4 GB VRAM –

Infinity-Parser2-Flashinfly · ~6 GB VRAM –

Infinity-Parser2-Proinfly · ~72 GB VRAM –

jina-ocr-v1Jina AI · ~8 GB VRAM –

Kimi K3Moonshot AI · ~1600 GB VRAM –

LightOnOCR-3 (4B, 1B, 0.8B)LightOn · ~11 GB VRAM –

Logics-Parsing-V3Alibaba (Logics-MLLM) · ~3 GB VRAM –

MonkeyOCRv2-B-Parsingzenosai (MonkeyOCR team) · ~3 GB VRAM –

Nanonets-OCR2-3BNanonets · ~9 GB VRAM –

olmOCR-2-7B-1025Allen Institute for AI (Ai2) · ~12 GB VRAM –

Qwen3.8-27BAlibaba (Qwen) · ~17 GB VRAM –

Surya OCR 2Datalab · ~3 GB VRAM –

WeVisDoc (2B and 4B)Tencent · ~10 GB VRAM

Apps

Cloud

Adobe Acrobat (AI Assistant)Adobe · $9.99/mo –

ChatGPTOpenAI · Free tier · $20/mo –

ChatPDFChatPDF · Free tier –

ClaudeAnthropic · Free tier · $20/mo –

Foxit PDF EditorFoxit · $11/mo –

GeminiGoogle · Free tier · $4.99/mo –

Gemini Notebook (formerly NotebookLM)Google · Free tier –

HumataHumata · Free tier · $9.99/mo –

MathpixMathpix · Free tier · $4.99/mo –

PDFgearPDFgear · Free –

SmallpdfSmallpdf · Free tier

Local

ABBYY FineReader PDFABBYY

Parsers & APIs

Cloud

Amazon TextractAmazon Web Services · Free tier –

anyformatanyformat · Free tier –

Azure Document IntelligenceMicrosoft · Free tier –

Document AIGoogle Cloud · Free tier –

ExtendExtend · Free tier · $500/mo –

LandingAI Agentic Document ExtractionLandingAI · Free tier · $250/mo –

LlamaParseLlamaIndex · Free tier · $50/mo –

MathpixMathpix · Free tier · $4.99/mo –

Nutrient Document Web ServicesNutrient · Free tier –

PulsePulse AI · Free tier –

ReductoReducto · Free tier –

Upstage Document ParseUpstage · Free tier

Local

PaddleOCRBaidu (PaddlePaddle) · Free$S MinerUOpenDataLab (Shanghai AI Laboratory) · Free –

MarkItDownMicrosoft · Free –

Tesseract OCRTesseract OCR project · Free – DoclingDocling Project (LF AI & Data) · Free –

MarkerDatalab · Free –

OCRmyPDFOCRmyPDF project · Free –

OpenDataLoader PDFOpenDataLoader project · Free –

UnstructuredUnstructured · Free –

LiteParseLlamaIndex · Free –

ChunkrLumina AI · Free –

PyMuPDF4LLMArtifex · Free

Full documents & ocr map, chart and table →

Embeddings & retrieval

Turn text into vectors and find what matters.

Best overallSharrier-oss-v1-27b
Best valueNo qualifying option yet
Best free startSOpenRouter
Best local (≤24 GB GPU)SKaLM-Embedding-Gemma3-12B-2511

Models

Cloud

Embed 5 ProCohere · $0.09/M tokens –

Gemini Embedding 2Google · $0.15/M tokens –

mistral-embed-2312Mistral AI · $0.08/M tokens –

Rerank 4 ProCohere –

text-embedding-3-largeOpenAI · $0.1/M tokens –

voyage-4-largeVoyage AI (MongoDB) · $0.09/M tokens

Local

harrier-oss-v1-27bMicrosoft · ~54 GB VRAM$S

KaLM-Embedding-Gemma3-12B-2511Tencent · ~24 GB VRAM$S

Qwen3-Embedding-8BAlibaba (Qwen) · ~16 GB VRAM$A

harrier-oss-v1-0.6bMicrosoft · ~2 GB VRAM$A

F2LLM-v2-14BAnt Group (codefuse-ai) · ~28 GB VRAM A

jina-embeddings-v5 (omni-small)Jina AI (Elastic) · ~4 GB VRAM B

nomic-embed-text-v2-moeNomic AI · ~2 GB VRAM B

snowflake-arctic-embed-l-v2.0Snowflake · ~3 GB VRAM B

zembed-1ZeroEntropy · ~8 GB VRAM C

granite-embedding-311m-multilingual-r2IBM · ~1 GB VRAM$–

bge-m3BAAI · ~3 GB VRAM –

bge-reranker-v2-m3BAAI · ~2 GB VRAM –

EmbeddingGemma 2Google · ~2 GB VRAM –

jina-reranker-v3.5Jina AI (Elastic) · ~2 GB VRAM –

Nemotron-3.5-Embed-8BNVIDIA · ~17 GB VRAM –

Qwen3-Reranker-8BAlibaba (Qwen) · ~17 GB VRAM –

zerank-2ZeroEntropy · ~9 GB VRAM

Retrieval frameworks

Cloud

Nothing tracked yet

Local

RAGFlowInfiniFlow · Free –

LlamaIndexLlamaIndex · Free –

Haystackdeepset · Free

Full embeddings & retrieval map, chart and table →