awembed is an open-source command-line tool for training compact embedding models tailored to a custom corpus. It combines teacher-model capture, distillation, fidelity-gated int8 quantization, and held-out evaluation for retrieval and code-search workflows.
In the Fine-tuning & training space, awembed takes a focused approach. It focuses on training efficient domain-specific embedding models for retrieval and code search without managing separate pipeline stages. awembed is an open-source project aimed at machine learning engineers and developers building custom retrieval systems. awembed is open source under the Apache-2.0 license. awembed is available on the command line, and it can be self-hosted.
Behind awembed is Aitherium, and it first shipped in 2026. Key capabilities include teacher capture, model distillation, and int8 quantization.
Summary written by a language model from the project’s public pages.
What PulseGate has recorded for this listing
Same category — not a similarity match