Nace AI有发布 Drex Max v1作为开放权重。 Drex Max是它最有能力的决定模型现在它运行在你的硬件上释放舰bf16重量，aQ8 0GGUF,affine Q4检查站和叉子llama.cpp和Ollama. 对于开发者来说，Drex Max 开放权重是指路由，分类和政策检查可以停留在本地机器上。
TL；数据
- 大小：Dense 27B参数，262,144个托肯窗口(默认为16,384个)，每个问题最多255个选项。
- 运行在：带Python的NVIDIA GPU(H100 bf16，或用8位的L40S 48 GB). Apple硅Macs配有24GB+(Affine Q4)或48GB(Q8 0 GGUF). CPU通过llama.cpp和Ollama.
- 业绩：关于公共决定指数0.4，比Drex v2.0(自报)高6.27分。
- 最佳：97.3 有关InfographyVQA的技能。 工具区达到81.9. 一个顶级的决策模型你可以运行离线在Mac。
Drex Max是什么来着？
Drex Max是奈斯公司的决定模型。 回答特定语境问题的AI. 您通过一个状态( 文本、 JSON 或图像) 加命名的问题。 模型返回一个前传中每个选项的概率。 无所生成。
问题分为3种：在命名选项中进行选择，是/否，以及一个正数分数。 这使得它适合agent的路由，票检和护栏检查。 主机版本在路由、分类和政策检查中介绍。
德雷克斯·麦克斯是怎么工作的？
模型卡描述了一个密集的27B型骨干，有64层。 48是Gated DeltaNet线性注意层，从左到右运行。 其他16个是全心全意的层层，每四层放置，它们都是无因子的。 每一个标志都可以关注整个序列。
图像通过一个本土的视觉塔，在请求中每个问题都可以看到。 一个255路读出头，然后将最后一个隐藏状态变成选项概率。 配置列出了qwen3 5型号，带有262,144位的窗口。
Drex Max在基准测试上表现如何？
以下所有数字均由Nace AI自行报告。
- 决策指数0.4(公开文本):67.52. Drex v2.0分 61.25分和Jev 1.13.0分57.96.
- 地区技能：工具81.9，语言72.2，检索68.8，知识和理性57.1，第53.4条。
——公共视觉指数：73.92对11基准测试，12,210行。
- 人工分析决定指数：64,Nace 称在18位列第4位。
- Jev Bench:92.2%(213/231)，裁判庭诉4:94.30%(1010/1071)。
结果数据集涵盖140 620项请求。 未答复的请求算错。
你怎么管理Drex Max本地？
Nace记录了5个运行时间。 它们都服务于相同的POST/v1/systemone API.
affine Q4检查站以24GB或以上的Macs为目标。 Q8 0 GGUF需要48GB. 腾讯网。 在48GB Mac上，Nace推荐–context 2048 for Kapitsa.
Ollama用户将 num gpu 40 设定为卸载40层到Metal. 设置 num gpu 0 在 CPU 上运行。 Ollama叉需要在模型文件中有一个Capability决定线。
在Nace的测试中，在48GB Apple M5 Pro上，Q8 0 GGUF在每概率0.02范围内匹配H100 bf16. 芳香Q4停留在Q8 0的0.0329之内，没有选择翻转。
Drex Max 与其他决定模型
Nace AI团队表示，决定指数0.4使用与0.3.1相同的公共套房。
关键外卖
- Drex Max v1重量现在在Nace下公开 AI Open RAI-M. (英语).
- 在决定指数0.4方面得分67.52，高于Drex v2.0(61.25)。
- 一个4位积分运行在24GB Apple硅Macs上，通过Kapitsa或MLX.
- M5 Pro上的Q8 0在0.02内与H100 bf16匹配。
- 本地Mac和CPU路径只是现在的文本。
请检查access-date=中的日期值 (帮助) 模型重量 Q8 0 GGUF，代码和docs,Tweet和云控制台。
感谢Nace AI团队为这篇文章提供的思想领导/资源。 这篇文章由Nace AI赞助。
Asif Razzaq是Marktechpost AI Media Inc.的首席执行官。 作为一个有远见的企业家和工程师，阿西夫致力于利用人工智能的潜力促进社会福祉。 他最近的努力是人工智能媒体平台的发布，该平台以深入报道机器学习和深层学习新闻而著称，这些新闻在技术上既可靠又为广大观众所容易理解。 该平台每月有超过200万次观看，表明其受观众欢迎。
本文由 AI图鉴自动抓取并翻译，版权归原作者所有；如需查看原文请点下方链接。
Nace AI has released Drex Max v1 as open weights. Drex Max is its most capable decision model, and it now runs on your own hardware. The release ships bf16 weights, a Q8_0 GGUF, an affine Q4 checkpoint and forks of llama.cpp and Ollama. For developers, Drex Max open weights mean routing, classification and policy checks can stay on a local machine.
TL;DR
- Size: Dense 27B parameters, 262,144-token window (16,384 by default), up to 255 options per question.
- Runs on: NVIDIA GPU with Python (H100 bf16, or L40S 48 GB in 8-bit). Apple Silicon Macs with 24 GB+ (affine Q4) or 48 GB (Q8_0 GGUF). CPU via llama.cpp and Ollama.
- Performance: 67.52 on the public Decision Index 0.4, 6.27 points ahead of Drex v2.0 (self-reported).
- Best: 97.3 skill on InfographicVQA. Tools area reaches 81.9. A top-tier decision model you can now run offline on a Mac.
What is Drex Max?
Drex Max is a decision model from Nace.AI that answers typed questions about a given context. You pass a state (text, JSON or images) plus named questions. The model returns a probability for every option in one forward pass. Nothing is generated.
Questions come in 3 types: choice among named options, yes/no, and an ordinal score. That makes it a fit for agent routing, ticket triage and guardrail checks. The hosted version is presented at routing, classification and policy checks.
How does Drex Max work?
The model card describes a dense 27B backbone with 64 layers. 48 are Gated DeltaNet linear-attention layers that run left to right. The other 16 are full-attention layers, placed every fourth layer, and they are non-causal. Every token can attend to the whole sequence.
Images pass through a native vision tower and are visible to every question in a request. A 255-way readout head then turns the last hidden state into option probabilities. The config lists a qwen3_5 model type with a 262,144-position window.
How does Drex Max perform on benchmarks?
All figures below are self-reported by Nace AI,
- Decision Index 0.4 (public text): 67.52. Drex v2.0 scores 61.25 and Jev 1.13.0 scores 57.96.
- Area skill: Tools 81.9, Language 72.2, Retrieval 68.8, Knowledge and Reasoning 57.1, Arts 53.4.
- Public vision index: 73.92 across 11 benchmarks and 12,210 rows.
- Artificial Analysis Decision Index: 64, which Nace says ranks 4th of 18.
- JevBench: 92.2% (213/231). Decision Bench v4: 94.30% (1010/1071).
The results dataset covers 140,620 requests. Unanswered requests count as wrong.
How do you run Drex Max locally?
Nace documents 5 runtimes. All of them serve the same POST /v1/systemone API.
The affine Q4 checkpoint targets Macs with 24 GB or more. The Q8_0 GGUF needs 48 GB. On a 48 GB Mac, Nace recommends –context 2048 for Kapitsa.
Ollama users set num_gpu 40 to offload 40 layers to Metal. Setting num_gpu 0 runs it on the CPU. The Ollama fork needs a CAPABILITY decision line in the Modelfile.
Quantization holds up in Nace’s tests. On a 48 GB Apple M5 Pro, the Q8_0 GGUF matched H100 bf16 within 0.02 per probability. The affine Q4 stayed within 0.0329 of Q8_0, with no choice flips.
Drex Max vs other decision models
Nace AI team states that Decision Index 0.4 uses the same public suite as 0.3.1.
Key Takeaways
- Drex Max v1 weights are now public under Nace.AI Open RAIL-M.
- It scores 67.52 on Decision Index 0.4, above Drex v2.0 (61.25).
- A 4-bit build runs on 24 GB Apple Silicon Macs via Kapitsa or MLX.
- Q8_0 on an M5 Pro matched H100 bf16 within 0.02.
- Local Mac and CPU paths are text only for now.
Check out the model weights, Q8_0 GGUF, code and docs, Tweet and cloud console.
Thanks to the Nace AI team for the thought leadership / resources for this article. This article is sponsored by Nace AI.
Asif Razzaq is the CEO of Marktechpost AI Media Inc.. As a visionary entrepreneur and engineer, Asif is committed to harnessing the potential of Artificial Intelligence for social good. His most recent endeavor is the launch of an Artificial Intelligence Media Platform, Marktechpost, which stands out for its in-depth coverage of machine learning and deep learning news that is both technically sound and easily understandable by a wide audience. The platform boasts of over 2 million monthly views, illustrating its popularity among audiences.
Auto-fetched and translated by AI Atlas. All rights belong to the original author.