Модели
2 канонические LLM-модели от всех провайдеров
Llama 4 Maverick1.0M ctx
Meta's quality-focused MoE model with 17B active parameters (400B total, 128 experts). Targets quality-critical tasks with benchmark scores competitive with GPT-4o and Gemini 2.5 Pro.
ЧатГенерация текстаВызов функций
Llama 4 Scout10.0M ctx
Meta's efficient MoE model with 17B active parameters (109B total, 16 experts). Supports up to 10M token context — the longest of any production model. Strong performance on reasoning and multilingual tasks.
ЧатГенерация текстаВызов функций