@QuixiAI: QuixiAI/embeddinggemma.c - fast as fuck cross-platform embedding inference engine. Written in c, runs anywhere. Inferen…

X AI KOLs Following Tools

Summary

QuixiAI releases embeddinggemma.c, a fast cross-platform embedding inference engine written in C, supporting multiple backends (CPU, Metal, CUDA, ROCm, SYCL) and Matryoshka embeddings with a standard HTTP API.

QuixiAI/embeddinggemma.c - fast as fuck cross-platform embedding inference engine. Written in c, runs anywhere. Inferences only embeddinggemma-300M-qat-q4_0-GGUF (2k context) Optimized kernels for cpu, metal, cuda, rocm, sycl. Matryoshka embeddings at 768, 512, 256, 128 dims. Standard http api. Did I mention that it's FAST?
Original Article
View Cached Full Text

Cached at: 07/21/26, 04:47 PM

QuixiAI/embeddinggemma.c - fast as fuck cross-platform embedding inference engine. Written in c, runs anywhere.

Inferences only embeddinggemma-300M-qat-q4_0-GGUF (2k context)

Optimized kernels for cpu, metal, cuda, rocm, sycl.

Matryoshka embeddings at 768, 512, 256, 128 dims.

Standard http api.

Did I mention that it’s FAST?

Similar Articles

zsxkib/embedding-gemma-300m

Replicate Explore

zsxkib/embedding-gemma-300m is a Replicate deployment of Google's EmbeddingGemma-300M model for generating 768-dimensional text embeddings, supporting flexible output dimensions via Matryoshka representation learning.

@QuixiAI: https://x.com/QuixiAI/status/2073936537213915611

X AI KOLs Following

QuixiAI released QuixiCore, a family of native high-performance AI kernel libraries for modern accelerators, with standalone implementations for CUDA, Metal, ROCm, XPU, and Gaudi backends, all sharing a common contract but no shared code.

Announcing Gemma 3n preview: Powerful, efficient, mobile-first AI

Google DeepMind Blog

Google announces Gemma 3n preview, a mobile-first open AI model optimized for on-device inference on phones, tablets, and laptops. Built on a new architecture developed with hardware partners like Qualcomm and MediaTek, Gemma 3n uses innovations like Per-Layer Embeddings to achieve fast performance with minimal memory footprint (2-3GB), while supporting multimodal capabilities.