NewMCP server: your agent can ask which model to useHow to connect →
live index / 481 models / updated daily
Match the model.
Skip the marketing.
Task-specific recommendations from public benchmarks and what developers actually report. Every claim comes with a receipt.
Start with the work
What are you building?
Every ranking
Best LLM for each task
Best LLM for Coding
Current #1: Claude Fable 5.2
- Best LLM for Agentic Coding · #1 Claude Fable 5.2
- Best LLM for Frontend & UI · #1 Claude Opus 5.5
- Best LLM for Code Completion · #1 Claude Sonnet 5.5
- Best LLM for Debugging & Code Review · #1 Claude Sonnet 5.5
- Best Local LLM for Coding · #1 DeepSeek V4.1 Flash 1017
Best LLM for AI Agents
Current #1: Claude Fable 5.2
- Best LLM for Tool & Function Calling · #1 Claude Fable 5.2
- Best LLM for Autonomous Agents · #1 Claude Fable 5.2
- Best LLM for Computer & Browser Use · #1 Claude Fable 5.2
- Best LLM for OpenClaw · #1 Claude Fable 5.2
- Best LLM for Hermes Agent · #1 Claude Fable 5.2
- Best LLM for Goose · #1 Claude Fable 5.2
- Best LLM for OpenHands · #1 Claude Fable 5.2
Best LLM for Writing
Current #1: Gemini 3.8 Flash
- Best LLM for Creative Writing · #1 Gemini 3.8 Flash
- Best LLM for Marketing Copy · #1 Gemini 3.8 Flash
- Best LLM for Business Writing · #1 Gemini 3.8 Flash
Best LLM for RAG & Long Context
Current #1: Gemini 4 Carbon
- Best LLM for Massive Context · #1 Gemini 4 Carbon
- Best LLM for Grounded RAG · #1 Gemini 4 Carbon
Best LLM for Chat & Roleplay
Current #1: Gemini 3.8 Flash
- Best LLM for a General Assistant · #1 Gemini 4 Carbon
- Best LLM for Roleplay & Character · #1 Gemini 3.8 Flash
Best LLM for Translation
Current #1: Gemini 3.8 Flash
- Best LLM for Everyday Translation · #1 Gemini 3.8 Flash
- Best LLM for Literary Translation · #1 Gemini 3.8 Flash
Best LLM for Summarization
Current #1: Gemini 4 Carbon
- Best LLM for Document Summarization · #1 Gemini 4 Carbon
- Best LLM for Fast Summaries · #1 Gemini 3.8 Flash
Best LLM for Structured Data Extraction
Current #1: DeepSeek V4.1 Flash
- Best LLM for JSON & Schema Output · #1 DeepSeek V4.1 Flash
- Best LLM for High-Volume Extraction · #1 GLM 5.2
- Best LLM for Document Parsing · #1 Gemini 3.8 Flash
Best LLM for Math & Reasoning
Current #1: Claude Opus 5.5
- Best LLM for Competition Math · #1 Claude Opus 5.5
- Best LLM for Logic & Problem Solving · #1 Gemini 4 Carbon
Best Cheap LLM for High Volume
Current #1: GLM 5.2
- Cheapest Capable LLM · #1 GLM 5.2
- Best Value LLM · #1 DeepSeek V4.1 Flash
- Best Open-Weight LLM for Self-Hosting · #1 GLM 5.2
Best LLM for Vision & Documents
Current #1: Gemini 3.8 Flash
- Best LLM for Image Understanding · #1 Claude Opus 5.5
- Best LLM for OCR & Documents · #1 Gemini 3.8 Flash
- Best LLM for Charts & Diagrams · #1 Gemini 3.8 Flash
Best Decision Model for Agent Routing and Scoring
Current #1: H2O-Lightning-4B v1.1 (H2O.ai, Qwen3.5-4B fine-tune, stock vLLM + open shim)
- Best Open-Weight Decision Model for Self-Hosting · #1 swanOne (blockbrain, Qwen3.8-Flash-Next NVFP4)
AI decision models for routing, classification and scoring
Choosing between Jev, an open-weight decision model and an LLM? Compare typed decisions, benchmark coverage and deployment options for your agent workflow.
Explore decision models and Jev alternatives →Recent model updates
Models change quickly. Our data follows.
New releases, price moves, and capability changes are logged as they land. Named variants such as Luna, Sol, and Terra are tracked as separate models.
View all updates →- $1.50 / M
- $1.13 / M
- Google: Gemini 4 Carbongoogle$2 / M
- Nex AGI: Nex-N2.6nex-agi$0.12 / M
- Anthropic: Claude Fable 5.2anthropic$8 / M
Question-specific
Coding agents and cheap extraction do not need the same leaderboard.
Evidence attached
Open a source beside any meaningful strength, weakness, or verdict.
Built to change
Recommendations refresh when benchmarks, prices, or credible reports move.