Open releases of sparse and mixture-of-experts architectures are reducing self-hosted inference costs and enabling distillation into smaller models. Projects such as Ling, Clover, and Nanbeige highlight open weights and base checkpoints as key enablers for fine-tuning and local deployment.
llama.cpp v0.8.0 adds compatibility with DFlash2 for Qwen3.8-Flash-Next. Tests on Apple Silicon M5 Max with quantized GGUFs show an average 17% decoding throughput gain over MTP, with peaks of 39% on list-like prompts and a minor prose regression. Th...
DiffuSpace has released Clover-35B-A3B-Preview, an open-source diffusion language model with a 35B MoE backbone, about 3B active parameters per token, and a 128K context window. The per-token active count, not total size, is the key metric for self-h...
The Ling-3.1-flash-1019 checkpoint is now on Hugging Face along with a BF16 base model for custom post-training. Total parameters remain 560B with 25B active and a 1M token context window. Reported scores move from 41 to 44 on AA Intelligence Index v...
inclusionAI updates Ling-3.1-flash: 560B total parameters, 25B active per token, and a 1M token context window. Agentic and scientific benchmarks improve. The base checkpoint is opened for research. The article analyzes implications for self-hosting ...
Nanbeige4.3-33B is not just a benchmark: it is a teacher for a distillation pipeline. For self-hosted inference, the signal is clear: cost shifts from deployment to distillation. The 3B can run on modest hardware, but the teacher requires resources. ...
Nanbeige presents the 33B model as stronger than Qwen3.8-27B on several benchmarks, but the core message is the distillation pipeline: it will feed compact versions such as a 3B. For self-hosted deployments, a 33B is VRAM-heavy; the path to a 3B chan...
