AI Model2026
Runs local conversational and agent-style assistants with improved multi-turn tool use and instruction following. Its Q4_K_M release targets roughly 4GB workloads and supports up to a 64K-token context.
AI Model2026
Provides non-uniform GGUF quantizations of DeepSeek-V4.1-Flash-1017, using GSQ and RCO to allocate precision per tensor under a size budget. The IQ2_XXS build matches the base model on the three reported benchmarks.
AI Model2026
Turns natural-language descriptions into runnable Three.js scene code, exploring how a small language model can generate structured interactive 3D environments as well as general-purpose code.