Product
Stratum
A community open-source project focused on deploying large-scale language models on consumer-grade hardware. Through quantization and MoE architectures it enables models with hundreds of billions of parameters to run under low GPU memory.
Timeline (last 90 days)
Oct 13
The output speed of 94 tokens per second was measured at a specific quantization precision and context length, and the actual experience will fluctuate with factors such as prompt length and concurrency.
Unverified50%