Niko1221/Strata
Local inference engine for running the Qwen3.8-Flash-Next 125B MoE model on consumer NVIDIA GPUs.
Collecting fresh signals — velocity needs a few days of history.
collecting data…
star history
Strata is a C++ inference engine that runs a 125-billion-parameter mixture-of-experts language model on a single NVIDIA GPU with 12-24 GB of VRAM plus 64 GB of system RAM. It provides a one-click installer for Windows and Linux, exposes OpenAI- and Anthropic-compatible HTTP APIs on localhost, and supports optional image input. The project ships quantized model variants and reports token throughput benchmarks for short chat and long-context workloads.
Frequently asked
- What is Niko1221/Strata?
- Local inference engine for running the Qwen3.8-Flash-Next 125B MoE model on consumer NVIDIA GPUs.
- Is Strata open source?
- Yes — Niko1221/Strata is open source, released under the MIT license.
- What language is Strata written in?
- Niko1221/Strata is primarily written in C++.
- How popular is Strata?
- Niko1221/Strata has 1.1k stars on GitHub.
- Where can I find Strata?
- Niko1221/Strata is on GitHub at https://github.com/Niko1221/Strata.