webAI · Reasoning LLM

TwIL-8B

webAI has released TwIL-8B, an 8.79B dense reasoning model for maths, code and instruction following, post-trained from IBM's Granite 4.2 8B. It ships in balanced and verbose variants, with a 4.98 GiB Q4_K_M GGUF for CPU or modest-VRAM GPU inference. All benchmark scores are self-reported: 98.2 on MATH-500 and 81.6 on LiveCodeBench, but it trails Qwen3.5-9B on GPQA Diamond. Weights are on Hugging Face under a non-commercial licence.