🏆 The AI Podium

The reference ranking of the best AI models, updated every month.

November 2026 edition

🏆 The AI Podium

Official ranking · November 2026

View methodology See full podium

Generalists

Best all-rounders

Gemini 4 Argon

Google·98/100·$2/$10Standard

Claude Opus 5.5-0

Anthropic·96/100·$4/$20Standard

Claude Fable 5.1-1

Anthropic·95/100·$10/$50Premium

Gemini 3.8 Flash

Google·94/100·$0.75/$3.75Low cost

Muse Spark 1.3-2

Meta·94/100·$1.25/$4.25Standard

Code Specialists

SWE-bench · Terminal-Bench · HumanEval

Claude Opus 5.5

Anthropic·86/100·$4/$20Standard

Claude Fable 5.1

Anthropic·86/100·$10/$50Premium

Claude Fable 5

Anthropic·84/100·$10/$50Premium

GPT-5.6 Sol-0

OpenAI·81/100·$4/$20Standard

Claude Opus 4.7

Anthropic·77/100·$5/$25Premium

Vision Specialists

Multimodal & image

Gemini 3.1 Pro+1

Google·78/100·$2/$12Standard

Claude Opus 4.8+3

Anthropic·69/100·$5/$25Premium

GPT-5.5+4

OpenAI·67/100·$5/$30Premium

Claude Opus 4.7+4

Anthropic·66/100·$5/$25Premium

GPT-5

OpenAI·63/100·$1.25/$10Standard

Multilingual Specialists

FR · EN · CN & more

Mistral Large 3

Mistral AI·81/100·$0.5/$1.5Low costapache-2.0

GLM-5+3

Zhipu AI·71/100·$1/$3.2Low costMIT

Gemini 3.1 Pro+4

Google·69/100·$2/$12Standard

GPT-5.5

OpenAI·66/100·$5/$30Premium

Kimi K2.6

Moonshot AI·64/100·$0.95/$4Low costmodified-mit

Best value for money

Generalists · capability vs cost

Gemini 3.8 Flash

Google·90/100·$0.75/$3.75Low cost

Claude Haiku 5.5

Anthropic·88/100·$0.1/$0.5Low cost

Gemini 3.6 Flash-0

Google·87/100·$0.75/$3.75Low cost

Gemini 4 Argon

Google·86/100·$2/$10Standard

Step 5 Preview

StepFun·85/100·$1/$2.7Low cost

Best value for money · Code

Most coding power per dollar

GPT-5.6 Luna

OpenAI·79/100·$0.2/$1.2Low cost

Gemini 3.6 Flash

Google·76/100·$0.75/$3.75Low cost

Kimi K2.6

Moonshot AI·74/100·$0.95/$4Low costmodified-mit

GLM-5

Zhipu AI·73/100·$1/$3.2Low costMIT

Gemini 3 Flash

Google·70/100·$0.5/$3Low cost

Open Source

Self-hostable, free license

Kimi K2.6

Moonshot AI·72/100·$0.95/$4Low costmodified-mit

Mistral Small 3+2

Mistral AI·63/100apache-2.0

Mistral Large 3+3

Mistral AI·60/100·$0.5/$1.5Low costapache-2.0

DeepSeek V3+4

DeepSeek·59/100mit

Apertus 1.5 8B

Swiss AI Initiative·58/100Apache-2.0

Local · Laptop

Light · < 16 GB RAM

Gemma 4 12B

Google·88/100apache-2.0

Qwen3 8B

Alibaba·79/100apache-2.0

Apertus 1.5 8B

Swiss AI Initiative·75/100Apache-2.0

Ministral 3 8B

Mistral AI·74/100apache-2.0

Granite 4.1 8B-2

IBM·73/100apache-2.0

Local · Desktop

~24 GB RAM

Qwen3.8 27B

Alibaba·92/100apache-2.0

Gemma 4 31B

Google·89/100apache-2.0

Qwen3.6 35B-A3B

Alibaba·89/100apache-2.0

Gemma 4 26B-A4B

Google·86/100apache-2.0

Qwen3.6 27B

Alibaba·84/100apache-2.0

Local · Workstation/Server

Big VRAM

DeepSeek V4.1-Flash

DeepSeek·93/100mit

GLM-5+2

Zhipu AI·87/100·$1/$3.2Low costMIT

GLM-5.2-1

Zhipu AI·86/100mit

Qwen3.5 397B-A17B

Alibaba·86/100apache-2.0

Kimi K2.7 Code

Moonshot AI·84/100modified-mit

🔬 How is this ranking made?

A transparent algorithm combines official benchmarks, pricing, context window and freshness to score each model.

View full methodology →

📅 Monthly archives

November 2026 (current)