flameox

CPU-only LLM inference engine in pure Rust — 4-bit quantized models, hand-written AVX2 kernels, speculative decoding, and a local HTTP/MCP server. No GPU, no Python runtime.

FLOPS Compute Intelligence publishes settlement-grade GPU + inference token price indices with cryptographic publication receipts. Indices update hourly; every value cites an inputs receipt your code can verify against the recompute audit substrate. Gives any MCP host (Claude Code, Cursor, Windsurf

Comprehensive, scalable ML inference architecture using Amazon EKS, leveraging Graviton processors for cost-effective CPU-based inference and GPU instances for accelerated inference. Guidance provides a complete end-to-end platform for deploying LLMs with agentic AI capabilities, including RAG and M