Nace AI 发布 27B Drex Max：只做选择不生成文本，决策成本仅 $0.06/千次
专门给 AI 智能体做决策的 27B 小模型，单次前向就出概率，成本比 Claude Opus 5.5 便宜约 170 倍，延迟 183ms，适合高频小决策场景。
Nace AI 发布 27B 参数的 Drex Max 模型，不生成任何 token，而是在单次前向传播中为给定选项各输出一个概率。在 Decision Bench v4 上，Drex Max 与 GPT-6 Sol 统计上打平。价格方面，Drex Max 每千次决策 $0.06、延迟 183 ms，对比 Claude Opus 5.5 的 $10.17 和约 3.5 秒。训练采用 RLAF 方法，奖励经核验的答案和推理评判模型的概率，并用 KL 惩罚限制对监督训练的偏移。局限是不带思维链或工具时，HLE 采样题正确率仅 21%，低于 Opus 的 56%。
This is a brilliant release from @NaceAI .
It goes after the costly part of AI agents, the thousands of small choices, with a 27B model that only picks and never writes.
- Nace AI’s 27B Drex Max statistically ties GPT-6 Sol on Decision Bench v4 without generating a token.
- It reads a context and question, then returns one probability per supplied option in a single forward pass.
- Its RLAF training rewards verified answers plus a reasoning judge’s probabilities, while a KL penalty limits drift from supervised training.
- Drex Max costs $0.06 per 1K decisions at 183 ms, against $10.17 and about 3.5 seconds for Claude Opus 5.5.
Without chain-of-thought or tools, it answers 21% of sampled HLE questions against 56% for Opus.
- TestingCatalog11-04 18:02原文
