unsloth/gemma-4-26B-A4B-it-GGUF · Hugging Face [web]
Provides a GGUF quantization of a 26B model with 4B active parameters, directly matching the reader's interest in mid-size MoE architectures for local inference. https://huggingface.co/unsloth/gemma-4-26B-A4B-it-GGUF (huggingface.co)
GitHub - THUDM/AgentBench: A Comprehensive Benchmark to Evaluate LLMs as Agents (ICLR'24) · GitHub [web]
A comprehensive benchmark for evaluating LLMs specifically in their capacity to function as agents, aligning with the reader's interest in agent frameworks and evaluation. https://github.com/THUDM/AgentBench (github.com)