EverestQ L1

Gold Verified

Publisher: EverestQ Engineering · Category: Reasoning

Parameters

70B

Context Window

128K

Default Size

42.5 GB

Downloads

142,500

General-purpose flagship reasoning model with multi-step chain-of-thought execution and 128K context window.

PULL MODEL

eq pull everestq/l1

RUN INFERENCE

eq run everestq/l1 "Explain Paged KV Cache block allocation"

Attention Architecture

Utilizes Grouped-Query Attention (GQA) with 16-token fixed Paged KV Cache block allocation for zero-fragmentation memory residency.

ONNX & EQC Compatible

Compiles directly through the EQC toolchain into standalone .eqx binary packages with fused SwiGLU kernels.

Multi-Hardware Support

Auto-detects CUDA RTX/A100/H100, Apple Silicon Metal Performance Shaders, or AVX-512 CPU execution backends.