Аналитика LLM моделей
Прозрачный лидерборд 346 моделей в стиле Artificial Analysis: SOTA-карточки, scatter-график цены vs интеллекта, полный лидерборд и AI-новости.
Лучшие модели в категориях
Данные собираются
Цена vs Интеллект
13 моделей · размер точки = размер контекста · наведите для деталей
Топ-30 моделей
| # | Модель | Провайдер | Score | Контекст | Ввод $/M | Вывод $/M |
|---|---|---|---|---|---|---|
| 1 | OpenAI: GPT-5.6 Luna Pro | OpenAI | 54.3 | 1.1M | $1.00 | $6.00 |
| 2 | Anthropic: Claude Fable Latest | ~anthropic | 51.4 | 1M | $10.00 | $50.00 |
| 3 | Anthropic: Claude Opus 4.8 (Fast) | Anthropic | 50.3 | 1M | $10.00 | $50.00 |
| 4 | Anthropic: Claude Sonnet 5 | Anthropic | 50.2 | 1M | $2.00 | $10.00 |
| 5 | ByteDance Seed: Seed-2.0-Lite | Bytedance-seed | 49.5 | 262K | $0.250 | $2.00 |
| 6 | Claude Mythos Preview UNRELEASED Anthropic | Claude | 49.4 | — | — | — |
| 7 | Qwen: Qwen3 Coder 480B A35B | Qwen | 48.5 | 1.0M | $0.300 | $1.00 |
| 8 | OpenAI: GPT-5.6 Luna | OpenAI | 47.3 | 1.1M | $1.00 | $6.00 |
| 9 | OpenAI: GPT-5.6 Terra Pro | OpenAI | 46.1 | 1.1M | $2.50 | $15.00 |
| 10 | ByteDance Seed: Seed-2.0-Mini | Bytedance-seed | 45.9 | 262K | $0.100 | $0.400 |
| 11 | Qwen: Qwen3 Coder 480B A35B (free) | Qwen | 45.8 | 1.0M | — | — |
| 12 | OpenAI: GPT-5.6 Terra | OpenAI | 45.3 | 1.1M | $2.50 | $15.00 |
| 13 | Anthropic: Claude Opus 4.8 | Anthropic | 45.2 | 1M | $5.00 | $25.00 |
| 14 | Anthropic: Claude Opus 4.7 (Fast) | Anthropic | 44.7 | 1M | $30.00 | $150.00 |
| 15 | OpenAI: GPT-5.6 Sol Pro | OpenAI | 44.5 | 1.1M | $5.00 | $30.00 |
| 16 | Meta: Llama 4 Scout | Meta | — | 10M | $0.100 | $0.300 |
| 17 | Auto Router | Openrouter | — | 2M | — | — |
| 18 | Pareto Code Router | Openrouter | — | 2M | — | — |
| 19 | xAI: Grok 4.20 | X-ai | — | 2M | $1.25 | $2.50 |
| 20 | xAI: Grok 4.20 Multi-Agent | X-ai | — | 2M | $1.25 | $2.50 |
| 21 | OpenAI: GPT-5.4 | OpenAI | — | 1.1M | $2.50 | $15.00 |
| 22 | OpenAI: GPT-5.4 Pro | OpenAI | — | 1.1M | $30.00 | $180.00 |
| 23 | OpenAI: GPT-5.5 | OpenAI | — | 1.1M | $5.00 | $30.00 |
| 24 | OpenAI: GPT-5.5 Pro | OpenAI | — | 1.1M | $30.00 | $180.00 |
| 25 | OpenAI: GPT-5.6 Sol | OpenAI | — | 1.1M | $5.00 | $30.00 |
| 26 | OpenAI GPT Latest | ~openai | — | 1.1M | $5.00 | $30.00 |
| 27 | Google: Gemini 3.1 Pro Preview Custom Tools | — | 1.0M | $2.00 | $12.00 | |
| 28 | DeepSeek: DeepSeek V4 Flash | DeepSeek | — | 1.0M | $0.098 | $0.196 |
| 29 | DeepSeek: DeepSeek V4 Pro | DeepSeek | — | 1.0M | $0.435 | $0.870 |
| 30 | Google: Gemini 2.5 Flash | — | 1.0M | $0.300 | $2.50 |
AI Новости
Kimi's open model K3 nears GPT-5.6 Sol and Fable 5 while signaling the end of super cheap Chinese AI
Kimi is launching K3, a multimodal open-weight model with 2.8 trillion parameters and one million tokens of context. In the company's own benchmarks, it comes close to Claude Fable 5 and GPT 5.6 Sol while beating Opus 4.8 and GLM 5.2, in some cases by a wide margin. The model is also signif
Читать →Google Vids now lets you star in your own AI videos
Читать →Roblox launches an AI-powered game-creation feature in its mobile app
Читать →Google rebrands NotebookLM as Gemini Notebook and opens its search app to third-party integration
Google is renaming NotebookLM to Gemini Notebook and integrating the tool more deeply into its ecosystem. A new feature gives each notebook its own cloud computer that can write and run code, initially for AI Ultra and Workspace customers. Separately, Google Search is getting app connection
Читать →Germany puts Google's AI Overviews and Perplexity under media law in first-of-its-kind ruling
German media regulators say Google's AI Overviews are Google's own content, not neutral search results, and that they crowd out regular links. The regulators have issued their first rulings against Google and Perplexity under the country's State Media Treaty. Both companies have one month t
Читать →О платформе LQD Intelligence
Intelligence — профессиональная аналитика LLM моделей, аналог Artificial Analysis и llm-stats.com. Здесь собраны 346 моделей от OpenAI, Anthropic, Google, Z.AI, DeepSeek, xAI, Mistral, Meta, Qwen и десятков других провайдеров.
Каждая модель представлена с ценой на ввод и вывод (USD за миллион токенов), размером контекста, модальностями, бенчмарками (GPQA, MMLU, HumanEval, Math) и скоростью. График «Цена vs Интеллект» помогает найти оптимальное соотношение цены и качества.
Все модели доступны через единый OpenAI-совместимый API Models Platform с оплатой в рублях.