webAI 开源 TwIL-8B：4.98 GiB 量化即可在 CPU 上跑
webAI 开源了 TwIL-8B，量化后不到 5 GiB 就能在 CPU 上跑，编程和数学成绩却逼近甚至超过多个前沿大模型，老电脑也能玩。
webAI 开源了基于 IBM Granite 4.2 8B 训练的 TwIL-8B，参数量 8.8B，主打数学、编程和指令跟随。量化文件只有 4.98 GiB，能在 CPU 或普通显卡上运行。它在 LiveCodeBench 拿到 81.6%，比 Qwen3.5-9B 高 16.0 分，也超过了 3.5 倍大的 Gemma 4 31B。对比 GPT-5.5、Claude Opus 4.8 和 Grok 4.3，它在编程达到最强前沿模型分数的 93%，并在 IFBench 上以 12.2 分优势击败 Claude Opus 4.8。得益于奖励更短推理的末段训练，它在 AIME 2026 上比 Qwen3.5-9B 少用 38% 的 token，MATH-500 得分 98.2% 为同组最高。
webAI (@thewebAI) just open-sourced TwIL-8B, an 8.8B model built on IBM’s Granite 4.2 8B and trained for math, coding and following instructions.
> TwIL-8B fits in a 4.98 GiB quantized file that runs on a CPU or modest graphics card, putting near-frontier coding and math on hardware people already own.
> scores 81.6% on LiveCodeBench, 16.0 points above Qwen3.5-9B and ahead of Google’s Gemma 4 31B, a model 3.5x its size.
> Against GPT-5.5, Claude Opus 4.8 and Grok 4.3, TwIL-8B reaches 93% of the best frontier score on coding and 92% on instruction following, beating Claude Opus 4.8 on IFBench by 12.2 points.
> A final training stage that rewards shorter reasoning lets TwIL-8B solve AIME 2026 problems with 38% fewer tokens than Qwen3.5-9B, while posting the highest MATH-500 score in its comparison at 98.2%.
