Upload 2 files

README.md

@@ -1,3 +1,453 @@
----
-license: mit
----
+<!-- markdownlint-disable first-line-h1 -->
+<!-- markdownlint-disable html -->
+<!-- markdownlint-disable no-duplicate-header -->
+
+<p align="left">
+  <a href="README_CN.md">中文</a>&nbsp;｜&nbsp;<b>English</b>
+</p>
+
+<div align="center">
+  <img src="assets/atria.png" width="60%" alt="Atria Dawn" />
+</div>
+
+<div align="center">
+
+# Atria Dawn
+
+**From Research Questions to Verifiable Results**
+
+</div>
+<div align="center" style="line-height: 1;">
+  <a href="https://huggingface.co/internlm/Atria-Dawn-Preview" target="_blank" style="margin: 2px;">
+    <img alt="Hugging Face" src="https://img.shields.io/badge/%F0%9F%A4%97%20Hugging%20Face-Atria%20Dawn%20Preview-FFD21E?labelColor=555" style="display: inline-block; vertical-align: middle;"/>
+  </a>
+  <a href="https://www.modelscope.cn/models/Shanghai_AI_Laboratory/Atria-Dawn-Preview" target="_blank" style="margin: 2px;">
+    <img alt="ModelScope" src="https://img.shields.io/badge/ModelScope-Atria%20Dawn%20Preview-624AFF?logo=ModelScope&logoColor=white&labelColor=555" style="display: inline-block; vertical-align: middle;"/>
+  </a>
+  <a href="LICENSE" style="margin: 2px;">
+    <img alt="License" src="https://img.shields.io/badge/License-MIT-3DA639?labelColor=555" style="display: inline-block; vertical-align: middle;"/>
+  </a>
+</div>
+
+<div align="center" style="line-height: 1;">
+  <a href="https://atria-asi.ai/" target="_blank" style="margin: 2px;">
+    <img alt="Website" src="https://img.shields.io/badge/Website-atria--asi.ai-1A73E8?logo=googlechrome&logoColor=white&labelColor=555" style="display: inline-block; vertical-align: middle;"/>
+  </a>
+  <a href="https://github.com/atria-asi/Atria-Dawn-Preview" target="_blank" style="margin: 2px;">
+    <img alt="GitHub" src="https://img.shields.io/badge/GitHub-Atria--Dawn--Preview-181717?logo=github&logoColor=white&labelColor=555" style="display: inline-block; vertical-align: middle;"/>
+  </a>
+  <a href="https://x.com/AtriaASI" target="_blank" style="margin: 2px;">
+    <img alt="Twitter Follow" src="https://img.shields.io/badge/Follow-%40AtriaASI-000000?logo=x&logoColor=white&labelColor=555" style="display: inline-block; vertical-align: middle;"/>
+  </a>
+  <a href="https://discord.gg/jT8SDt8up" target="_blank" style="margin: 2px;">
+    <img alt="Discord" src="https://img.shields.io/badge/Discord-Join%20Chat-5865F2?logo=discord&logoColor=white&labelColor=555" style="display: inline-block; vertical-align: middle;"/>
+  </a>
+</div>
+
+<div align="center"  style="line-height: 1;">
+  <a href="https://arxiv.org/abs/2609.15818" target="_blank" style="margin: 2px;">
+    <img alt="arXiv" src="https://img.shields.io/badge/Paper-Atria Dawn-b31b1b?logo=arxiv&logoColor=red&labelColor=555" style="display: inline-block; vertical-align: middle;"/>
+  </a>
+</div>
+
+<br>
+
+<div align="center">
+  <a href="#-introduction">Introduction</a> ·
+  <a href="#-model-downloads">Model Downloads</a> ·
+  <a href="#-evaluation-results">Evaluation</a> ·
+  <a href="#-deployment--online-access">Deployment</a> ·
+  <a href="#-license">License</a> ·
+  <a href="#-contact">Contact</a>
+</div>
+
+<hr>
+
+## 🌅 Introduction
+
+**Atria Dawn** is a new-generation agentic model developed by the **Shanghai Artificial Intelligence Laboratory**. Built on the **744B-parameter MoE GLM-5.2** foundation model, it targets research and engineering scenarios that demand continuous environmental understanding, tool use, and multi-step task completion.
+
+The model is designed to drive open-ended problems toward **executable, verifiable, and reproducible** results. By combining task objectives with environmental feedback, it supports the full loop of problem analysis, solution design, tool use, code implementation, experiment execution, result analysis, and failure recovery.
+
+Atria Dawn empowers agentic tasks across four dimensions, with a particular focus on end-to-end delivery in real-world productivity scenarios such as scientific automation and office work:
+
+<div align="center">
+
+| | Dimension | What it covers |
+| :---: | :--- | :--- |
+| 🔍 | **Discovery** | Retrieving and organizing evidence, conducting deep research, and turning research questions into executable experimental plans. |
+| 🛠️ | **Creation** | Building software, interactive applications, games, data visualizations, and machine learning systems. |
+| 📦 | **Delivery** | Transforming documents, data, and design requirements into reports, presentations, and other structured deliverables. |
+| 🛡️ | **Cybersecurity** | Analyzing security issues, validating vulnerabilities, applying fixes, and performing re-validation in authorized environments. |
+
+</div>
+
+## 📥 Model Downloads
+
+<div align="center">
+
+| Model | Description | Context | Hugging Face | ModelScope |
+| :--- | :--- | :---: | :---: | :---: |
+| **Atria Dawn** | Instruct model (full release) | 512K | Coming soon | Coming soon |
+| **Atria Dawn FP8** | FP8-quantized Instruct model (full release) | 512K | Coming soon | Coming soon |
+| **Atria Dawn Preview** | Instruct model | 256K | [🤗 Model](https://huggingface.co/internlm/Atria-Dawn-Preview) | [🔗 Model](https://www.modelscope.cn/models/Shanghai_AI_Laboratory/Atria-Dawn-Preview) |
+| **Atria Dawn Preview FP8** | FP8-quantized Instruct model | 256K | [🤗 Model](https://huggingface.co/internlm/Atria-Dawn-Preview-FP8) | [🔗 Model](https://www.modelscope.cn/models/Shanghai_AI_Laboratory/Atria-Dawn-Preview-FP8) |
+
+</div>
+
+The model name is **Atria Dawn**. Use `Atria-Dawn` as the `model` value in API requests.
+
+## 📊 Evaluation Results
+
+We evaluated Atria Dawn and Atria Dawn Preview across search, coding, tool use, productivity, and security benchmarks. The full table below matches the [official website](https://atria-asi.ai/).
+
+<div align="center">
+  <img src="assets/evaluation.jpg" width="100%" alt="Atria Dawn and Atria Dawn Preview Evaluation Results" />
+</div>
+
+<div align="center">
+
+| Benchmark | Atria Dawn | Atria Dawn Preview | DeepSeek V4 Pro 0813 | KIMI K3 | Qwen 3.8 Max | GLM 5.3 | GPT 5.6 sol | Claude Opus 5 |
+| :--- | :---: | :---: | :---: | :---: | :---: | :---: | :---: | :---: |
+| AutomationBench | **54.7** | 53.8 | 41.7 | 45.9 | 49.7 | 52.5 | 45.7 | 49.4 |
+| CyberGym | **86.8** | 86.5 | 83.3 | 78.7 | 73.8 | 84.5 | 83.6 | – |
+| SkillsBench | **67.8** | 66.4 | 65.0 | 51.9 | 66.7 | 63.3 | 62.5 | 63.7 |
+| WideSearch | **84.6** | 81.9 | 81.8 | 79.6 | 81.9 | 82.7 | 83.3 | 84.0 |
+| MLE-bench Lite | 88.2 | 86.2 | 86.8 | 85.8 | 81.3 | 80.8 | **88.9** | 88.0 |
+| Workspace-Bench-Lite | 70.0 | 68.2 | 62.1 | 64.9 | 68.7 | **71.7** | 60.5 | 70.1 |
+| Workspace-Bench | 66.1 | 65.0 | 58.2 | 63.5 | **68.0** | 67.4 | 56.0 | 65.8 |
+| PaperBench | 89.6 | 87.0 | 87.4 | 80.5 | **94.1** | 86.9 | 92.8 | 88.9 |
+| JobBench | 57.5 | 50.3 | 54.1 | 54.3 | 52.7 | 58.2 | 45.4 | **68.0** |
+| GDPval | 1594.2 | 1583.8 | 1493.3 | 1606.8 | 1710.9 | 1643.1 | 1611.1 | **1714.0** |
+| DeepResearch Bench II | 51.2 | 51.1 | 46.6 | 51.3 | 49.2 | 52.7 | 50.7 | **54.1** |
+| SpreadsheetBench 2 | 33.0 | 31.5 | 29.3 | **35.8** | 34.9 | 32.1 | 32.7 | 33.6 |
+| τ³-bench Banking | 43.3 | 41.2 | 44.3 | 37.1 | **55.2** | 40.2 | 46.9 | 48.7 |
+| Terminal-Bench 2.1 | 81.0 | 78.3 | 77.0 | 84.0 | 89.6 | 86.0 | 86.2 | **90.2** |
+| SWE-bench Pro | 61.7 | 59.6 | 58.4 | 62.0 | 65.3 | 60.3 | 59.8 | **76.5** |
+| NL2Repo-Bench | 50.0 | 51.4 | 49.5 | 63.1 | 63.2 | 54.8 | 60.7 | **79.9** |
+| Harbor-Index | 17.3 | 17.3 | 17.0 | 24.3 | 25.8 | 20.0 | 35.8 | **41.0** |
+| DeepSWE | 49.3 | – | 54.9 | 70.6 | 67.0 | 67.3 | 70.6 | **72.6** |
+| Harvey LAB | **27.5** | – | 6.7 | 25.4 | 24.2 | 22.5 | 5.0 | 22.7 |
+| BFCL V4 | **78.4** | 77.0 | 71.4 | 69.1 | – | 74.1 | – | – |
+| DeepSearchQA | **96.5** | 96.0 | – | 95.9 | – | 94.7 | 93.2 | – |
+| OfficeQA Pro | 67.1 | – | 66.2 | 65.4 | 66.2 | 65.8 | 66.4 | **67.2** |
+| BrowseComp | 91.3 | **92.5** | 83.4 | 91.2 | – | 89.6 | 92.2 | 90.8 |
+
+</div>
+
+> **Bold** marks the best score in each row. `–` indicates the result is unavailable.
+
+<!-- ## How to Run Locally
+
+Add instructions for local inference, deployment, and environment setup here.
+
+```bash
+# Add installation and launch commands here.
+``` -->
+
+## 🚀 Deployment & Online Access
+
+Atria Dawn supports both local deployment and hosted access.
+
+### Online Access
+
+Use the service endpoint corresponding to your region.
+
+<div align="center">
+
+| Region | Access | Tutorial |
+| :--- | :--- | :--- |
+| 🌐 **International** | [API Console](https://api.atria-asi.ai/) | [Docs](https://api.atria-asi.ai/docs) |
+| 🇨🇳 **China** | [API Console](https://discovery.intern-ai.org.cn/token-plan/home?tabIndex=0) | [Docs](https://discovery.intern-ai.org.cn/token-plan/home?tabIndex=3) |
+
+</div>
+
+### Local Deployment
+
+<div align="center">
+
+| Framework | Minimum Version | Guide |
+| :--- | :---: | :--- |
+| [**SGLang**](https://github.com/sgl-project/sglang) | `v0.5.13.post1+` | [Cookbook](https://cookbook.sglang.io/autoregressive/GLM/GLM-5.2) |
+| [**vLLM**](https://github.com/vllm-project/vllm) | `v0.23.0+` | [Recipes](https://recipes.vllm.ai/zai-org/GLM-5.2) |
+
+</div>
+
+### Codex
+Add a custom provider to `~/.codex/config.toml`. Codex uses the Responses API.
+
+```toml
+model = "Atria-Dawn"
+model_provider = "atria"
+[model_providers.atria]
+name = "Atria"
+base_url = "https://api.atria-asi.ai/v1"
+env_key = "ATRIA_API_KEY"
+wire_api = "responses"
+```
+
+#### Restricting input to text only
+
+**Atria Dawn** accepts text input only. By default Codex assumes every model is
+multimodal and will attach images from `-i/--image` or a TUI paste, which the endpoint
+rejects with `400 Atria-Dawn is not a multimodal model`. Declare the model's
+modalities so Codex strips image input on the client side instead.
+
+##### Step 1 — Create a model catalog file
+
+Save this as `~/.codex/atria-catalog.json`:
+
+```json
+{
+  "models": [
+    {
+      "slug": "Atria-Dawn",
+      "display_name": "Atria Dawn",
+      "base_instructions": "You are a coding agent running in the Codex CLI. You collaborate with the user in a shared workspace to accomplish their software engineering goals.\n\nYou can only receive text input. Images, screenshots, PDFs, and other binary attachments are not available to you. If the user refers to an attachment you cannot see, say so plainly and ask them to paste the relevant text instead.",
+      "supported_reasoning_levels": [
+        { "effort": "low", "description": "Fast responses with lighter reasoning" },
+        { "effort": "medium", "description": "Balances speed and reasoning depth" },
+        { "effort": "high", "description": "Greater reasoning depth for complex problems" }
+      ],
+      "shell_type": "unified_exec",
+      "visibility": "list",
+      "supported_in_api": true,
+      "priority": 1,
+      "support_verbosity": false,
+      "truncation_policy": { "mode": "tokens", "limit": 10000 },
+      "experimental_supported_tools": [],
+      "context_window": 256000,
+      "max_context_window": 256000,
+      "input_modalities": ["text"]
+    }
+  ]
+}
+```
+
+`"input_modalities": ["text"]` is the setting that disables multimodal input.
+
+Set `context_window` / `max_context_window` to the model's real limit — Codex uses these
+to budget the prompt and decide when to auto-compact. Without them it falls back to a
+conservative default, which wastes usable context.
+
+All other fields are required by the parser — omitting any one fails with
+`missing field <name>` and Codex will not start.
+
+##### Step 2 — Point your config at it
+
+```toml
+model = "Atria-Dawn"
+model_provider = "atria"
+model_catalog_json = "~/.codex/atria-catalog.json"
+[features]
+view_image = false
+[model_providers.atria]
+name = "Atria"
+base_url = "https://api.atria-asi.ai/v1"
+env_key = "ATRIA_API_KEY"
+wire_api = "responses"
+```
+
+`features.view_image = false` is optional — it removes the image-viewing tool so the model
+doesn't attempt a call that would be refused.
+
+> **Important:** `model_catalog_json` **replaces** the model catalog, it does not merge
+> with it. Any model not listed in your file falls back to default metadata — which
+> assumes multimodal input — and logs
+> `warning: Model metadata for <slug> not found`. If you switch models with `-m` or by
+> editing `model`, add that model to the same file, or the text-only restriction will not
+> apply to it.
+
+Requires Codex CLI 0.154.0 or later.
+
+### Claude Code
+```Python
+#!/usr/bin/env python3
+"""
+PreToolUse hook: block the Read tool from reading PDF and image files.
+"""
+import json
+import sys
+IMAGE_EXTENSIONS = (
+    ".apng",
+    ".avif",
+    ".bmp",
+    ".gif",
+    ".heic",
+    ".heif",
+    ".ico",
+    ".jfif",
+    ".jpeg",
+    ".jpg",
+    ".jxl",
+    ".png",
+    ".svg",
+    ".tif",
+    ".tiff",
+    ".webp",
+)
+def main():
+    hook_input = json.loads(sys.stdin.read())
+    file_path = hook_input.get("tool_input", {}).get("file_path", "")
+    if file_path.lower().endswith(".pdf"):
+        reason = "Reading PDF files with the Read tool is not allowed."
+    elif file_path.lower().endswith(IMAGE_EXTENSIONS):
+        reason = "Reading image files with the Read tool is not allowed."
+    else:
+        sys.exit(0)
+    output = {
+        "hookSpecificOutput": {
+            "hookEventName": "PreToolUse",
+            "permissionDecision": "deny",
+            "permissionDecisionReason": reason,
+        }
+    }
+    print(json.dumps(output, ensure_ascii=False))
+    sys.exit(0)
+if __name__ == "__main__":
+    main()
+```
+
+Please save the above code script as ${Target_dir}/block_pdf_image_read.py, and be sure to use an absolute path.
+
+Add it to ~/claude_dir/settings.json.
+This can implement interception of multimodal inputs, such as images and PDFs, in PreToolUse.
+```Bash
+"hooks": {
+    "PreToolUse": [
+      {
+        "matcher": "Read",
+        "hooks": [
+          {
+            "type": "command",
+            "command": "python3 ${Target_dir}/block_pdf_image_read.py",
+            "timeout": 5
+          }
+        ]
+      },
+      {
+        "matcher": "",
+        "hooks": []
+      }
+    ]
+}
+```
+### Kimi Code
+
+Add a custom provider to `~/.kimi-code/config.toml`. Kimi Code talks to Atria over the OpenAI-compatible Chat Completions API.
+
+This example targets Kimi Code 0.43.1 and later. Check your version first; config fields can vary between versions.
+
+```bash
+kimi --version
+cp -a ~/.kimi-code/config.toml ~/.kimi-code/config.toml.bak
+```
+
+#### 1. Write the provider and model into your user config
+
+Merge this into `~/.kimi-code/config.toml`. Keep `default_model` at the top level, before the tables. Update existing entries instead of duplicating fields or tables. Your `atr_...` key goes on the `api_key` line — keys are created in the Atria console and displayed only once.
+
+```toml
+default_model = "atria/Atria-Dawn"
+
+[providers.atria]
+type = "openai"
+# must include /v1 — Kimi appends /chat/completions verbatim
+base_url = "https://api.atria-asi.ai/v1"
+# literal value only: no ${VAR} expansion, and OPENAI_API_KEY is not read
+api_key = "ATRIA_API_KEY"
+
+[models."atria/Atria-Dawn"]
+provider = "atria"
+# case-sensitive wire name, separate from the section key above (a local alias)
+model = "Atria-Dawn"
+display_name = "Atria Dawn"
+# required — without it the entry silently fails to load. Keep at the real limit
+# so context budgeting and compaction work.
+max_context_size = 256000
+# the max_tokens sent on the wire. Without it Kimi sends max_context_size
+# instead, which Atria rejects (valid range is 1-65536).
+max_output_size = 65536
+# text-only model: adding image_in fails with "is not a multimodal model".
+# Declare tool_use explicitly, or Kimi treats capabilities as unknown.
+# "thinking" is what turns reasoning on — without it the effort is forced to
+# "off" and no reasoning_effort is sent.
+capabilities = ["tool_use", "thinking"]
+# without support_efforts the effort stays the opaque "on", which sends no
+# reasoning_effort at all. Listing them makes /think and the picker work.
+support_efforts = ["low", "medium", "high", "xhigh", "max"]
+default_effort = "max"
+# required to allow turning thinking off. Omitting it fails with "reasons by
+# default but declares no off effort".
+off_effort = "none"
+```
+
+Keep the file private, since the key is stored in plaintext:
+
+```bash
+chmod 600 ~/.kimi-code/config.toml
+```
+
+#### 2. Restart and verify
+
+Validate the config, then send a text-only task.
+
+```bash
+kimi doctor
+kimi provider list
+kimi
+```
+
+Expected:
+
+```
+OK config.toml  /home/<user>/.kimi-code/config.toml
+All checked config files are valid.
+
+atria  type=openai  models=1  source=inline
+Default model: atria/Atria-Dawn
+```
+
+If `atria` is missing from the provider list, the model entry failed to resolve — most often an absent `max_context_size`.
+
+
+## 📄 License
+
+The code and model weights in this repository are released under the [MIT License](LICENSE).
+
+<!-- ## Citation
+
+```text
+Add the official citation here.
+``` -->
+
+## 📮 Contact
+
+Questions, suggestions, or collaboration ideas are welcome:
+
+- 🐛 **Issues & feature requests** — open an issue on [GitHub](https://github.com/atria-asi/Atria-Dawn-Preview/issues)
+- 💬 **Community** — join us on [Discord](https://discord.gg/jT8SDt8up)
+- 📱 **WeChat** — scan the QR code below to join Atria community group 7
+- 🐦 **Updates** — follow [@AtriaASI](https://x.com/AtriaASI) on X
+
+<p align="center">
+  <img src="assets/wechat-community-7.png" alt="Atria WeChat community group 7 QR code" width="320" />
+</p>
+
+## Citation
+
+```
+@misc{guo2026atriadawndawnagentic,
+      title={Atria Dawn: The Dawn of Agentic Superintelligence}, 
+      author={Honglin Guo and Tao Gui and Yicheng Chen and Guanting Dong and Qiming Ge and Yuyang Hu and Zixian Huang and Jiajie Jin and Alexander Lam and Yining Li and Jiahang Lin and Yanjiang Liu and Xinyu Lu and Haijun Lv and Junlin Shang and Qisheng Su and Guoqiang Wang and Rui Wang and Zhecan Wang and Hao Xiang and Xinchen Xie and Shuhao Xing and Xiaoyu Xing and Wanghan Xu and Xinyu Yang and Yajie Yang and Chengfeng Zhao and Haoran Zhao and Ruojun Zhou and Yunhua Zhou and Yicheng Zou and Kun Cai and Qiye Cai and Xinmeng Che and Haodong Chen and Jiabei Chen and Jiahao Chen and Jiayi Chen and Yujia Chen and Lizhi Cui and Youheng Dai and Xin Deng and Yi Dong and Shihan Dou and Chenya Gu and Xu Guo and Ding Han and Feiyang Hao and Haotan He and Jie Hou and Binze Hu and Zijian Hu and Junhao Huang and Huicheng Jiang and Jiazhen Jiang and Shufan Jiang and Jiahao Kuang and Bowen Lai and Bo Li and Jiaqiang Li and Peng Li and Qilong Li and Zhuoqun Li and Jiaxiang Liu and Shuainan Liu and Tong Liu and Yi Liu and Zhonghang Lu and Jianwen Luo and Yanyi Luo and Huijie Lv and Ningsheng Ma and Zerun Ma and Houcheng Min and Chengjun Pan and Qiyuan Peng and Xiaoxuan Peng and Jianmin Qian and Jiantao Qiu and Wanying Ren and Huayu Sha and Jifei Shan and Zixin Shang and Bing Shao and Zhuohui Sheng and Jiayang Shi and Yang Shu and Aierpanjiang Simayi and Sirui Song and Yuxiao Song and Zhe Sun and Zhichao Sun and Wenzhe Tan and Wenhui Tian and Zhongbo Tian and Hanchen Wang and Pengbo Wang and Rui Wang and Yiding Wang and Yuhui Wang and Zhiheng Xi and Caijun Xu and Chao Xu and Yongfeng Xu and Xiaolei Yang and Zhixiong Yang and Qian Yao and Shihong Yi and Yuankai Ying and Jia Yu and Dingbo Yuan and Hao Yuan and Junjie Yuan and Bo Zhang and Caixian Zhang and Qiuyinzhe Zhang and Jiyuan Zhao and Penghao Zhao and Ying Zhao and Pujun Zheng and Xiaoxue Zhong and Xiaohao Zhou and Xinyu Zhou and Dongsheng Zhu and Guanru Zhu and Yulun Zhu and Yaojie Lu and Tao Ji and Hongyu Lin and Yutao Zhu and Pengfei Cao and Guoxiu He and Xianpei Han and Ben He and Zhicheng Dou and Kang Liu and Qi Zhang and Le Sun and Jun Zhao and Ji-Rong Wen and Xuanjing Huang and Yu-Gang Jiang and Bowen Zhou},
+      year={2026},
+      eprint={2609.15818},
+      archivePrefix={arXiv},
+      primaryClass={cs.AI},
+      url={https://arxiv.org/abs/2609.15818},
+      doi={10.48550/ARXIV.2609.15818},
+}
+```
+<div align="center">
+  <sub>Built by the Shanghai Artificial Intelligence Laboratory.</sub>
+</div>

README_CN.md

@@ -0,0 +1,427 @@
+<p align="left">
+   <a href="README.md">English</a>&nbsp;｜&nbsp;中文
+</p>
+<br>
+
+<div align="center">
+  <img src="assets/atria.png" width="60%" alt="Atria Dawn" />
+</div>
+
+<div align="center">
+
+# Atria Dawn: 从研究问题到可验证的结果
+
+<!-- markdownlint-disable first-line-h1 -->
+<!-- markdownlint-disable html -->
+<!-- markdownlint-disable no-duplicate-header -->
+
+</div>
+<div align="center" style="line-height: 1;">
+  <a href="https://huggingface.co/internlm/Atria-Dawn-Preview" target="_blank" style="margin: 2px;">
+    <img alt="Hugging Face" src="https://img.shields.io/badge/%F0%9F%A4%97%20Hugging%20Face-Atria%20Dawn%20Preview-FFD21E?labelColor=555" style="display: inline-block; vertical-align: middle;"/>
+  </a>
+  <a href="https://www.modelscope.cn/models/Shanghai_AI_Laboratory/Atria-Dawn-Preview" target="_blank" style="margin: 2px;">
+    <img alt="ModelScope" src="https://img.shields.io/badge/ModelScope-Atria%20Dawn%20Preview-624AFF?logo=ModelScope&logoColor=white&labelColor=555" style="display: inline-block; vertical-align: middle;"/>
+  </a>
+  <a href="LICENSE" style="margin: 2px;">
+    <img alt="License" src="https://img.shields.io/badge/License-MIT-3DA639?labelColor=555" style="display: inline-block; vertical-align: middle;"/>
+  </a>
+</div>
+
+<div align="center" style="line-height: 1;">
+  <a href="https://atria-asi.ai/" target="_blank" style="margin: 2px;">
+    <img alt="Website" src="https://img.shields.io/badge/Website-atria--asi.ai-1A73E8?logo=googlechrome&logoColor=white&labelColor=555" style="display: inline-block; vertical-align: middle;"/>
+  </a>
+  <a href="https://github.com/atria-asi/Atria-Dawn-Preview" target="_blank" style="margin: 2px;">
+    <img alt="GitHub" src="https://img.shields.io/badge/GitHub-Atria--Dawn--Preview-181717?logo=github&logoColor=white&labelColor=555" style="display: inline-block; vertical-align: middle;"/>
+  </a>
+  <a href="https://x.com/AtriaASI" target="_blank" style="margin: 2px;">
+    <img alt="Twitter Follow" src="https://img.shields.io/badge/Follow-%40AtriaASI-000000?logo=x&logoColor=white&labelColor=555" style="display: inline-block; vertical-align: middle;"/>
+  </a>
+  <a href="https://discord.gg/jT8SDt8up" target="_blank" style="margin: 2px;">
+    <img alt="Discord" src="https://img.shields.io/badge/Discord-Join%20Chat-5865F2?logo=discord&logoColor=white&labelColor=555" style="display: inline-block; vertical-align: middle;"/>
+  </a>
+</div>
+
+<div align="center"  style="line-height: 1;">
+  <a href="https://arxiv.org/abs/2609.15818" target="_blank" style="margin: 2px;">
+    <img alt="arXiv" src="https://img.shields.io/badge/Paper-Atria Dawn-b31b1b?logo=arxiv&logoColor=red&labelColor=555" style="display: inline-block; vertical-align: middle;"/>
+  </a>
+</div>
+
+<br>
+
+<div align="center">
+  <a href="#介绍">介绍</a> ·
+  <a href="#模型下载">模型下载</a> ·
+  <a href="#性能评估">性能评估</a> ·
+  <a href="#部署和在线使用">部署和在线使用</a> ·
+  <a href="#许可证">许可证</a> ·
+  <a href="#联系我们">联系我们</a> ·
+  <a href="#引用">引用</a>
+</div>
+
+## 介绍
+
+Atria Dawn 是由上海人工智能实验室研发的新一代智能体大模型，模型基于 744B MoE GLM-5.2 基座进行训练，面向需要持续理解环境、调用工具并完成多步任务的研究与工程场景。该模型旨在协助用户将开放性问题推进为可执行、可验证和可复现的结果。模型能够结合任务目标与环境反馈，参与问题分析、方案设计、工具调用、代码实现、实验执行、结果分析以及失败恢复等环节。
+
+Atria Dawn 从以下四个维度赋能智能体任务，重点强化了在科研自动化、办公等真实生产力场景下的端到端交付能力。
+
+- **Discovery：** 检索和组织证据，开展深度研究，并将研究问题转化为可执行的实验方案。
+- **Creation：** 构建软件、交互应用、游戏、数据可视化和机器学习系统。
+- **Delivery：** 将文档、数据和设计要求转化为报告、演示文稿及其他结构化成果。
+- **Cybersecurity：** 在经过授权的环境中分析安全问题、验证漏洞、实施修复并完成复验。
+
+
+
+## 模型下载
+
+<div align="center">
+
+| **模型名** | **简介** | **上下文** | **Hugging Face** | **ModelScope** |
+| :---: | :---: | :---: | :---: | :---: |
+| Atria Dawn | Instruct 模型（正式版） | 256K | Coming soon | Coming soon |
+| Atria Dawn FP8 | FP8量化Instruct模型（正式版） | 256K | Coming soon | Coming soon |
+| Atria Dawn Preview | Instruct 模型 | 256K | [Model](https://huggingface.co/internlm/Atria-Dawn-Preview) | [Model](https://www.modelscope.cn/models/Shanghai_AI_Laboratory/Atria-Dawn-Preview) |
+| Atria Dawn Preview FP8 | FP8量化Instruct模型 | 256K| [Model](https://huggingface.co/internlm/Atria-Dawn-Preview-FP8) | [Model](https://www.modelscope.cn/models/Shanghai_AI_Laboratory/Atria-Dawn-Preview-FP8) |
+
+</div>
+
+模型名称为 **Atria Dawn**；API 请求的 `model` 参数使用 `Atria-Dawn`。
+
+## 性能评估
+我们对 Atria Dawn 和 Atria Dawn Preview 进行了全面评估。以下完整大表与[官网](https://atria-asi.ai/)同步。
+
+
+<div align="center">
+  <img src="assets/evaluation.jpg" width="100%" alt="Atria Dawn and Atria Dawn Preview Evaluation Results" />
+</div>
+
+<div align="center">
+
+| 评测项目 | Atria Dawn | Atria Dawn Preview | DeepSeek V4 Pro 0813 | KIMI K3 | Qwen 3.8 Max | GLM 5.3 | GPT 5.6 sol | Claude Opus 5 |
+| :--- | :---: | :---: | :---: | :---: | :---: | :---: | :---: | :---: |
+| AutomationBench | **54.7** | 53.8 | 41.7 | 45.9 | 49.7 | 52.5 | 45.7 | 49.4 |
+| CyberGym | **86.8** | 86.5 | 83.3 | 78.7 | 73.8 | 84.5 | 83.6 | – |
+| SkillsBench | **67.8** | 66.4 | 65.0 | 51.9 | 66.7 | 63.3 | 62.5 | 63.7 |
+| WideSearch | **84.6** | 81.9 | 81.8 | 79.6 | 81.9 | 82.7 | 83.3 | 84.0 |
+| MLE-bench Lite | 88.2 | 86.2 | 86.8 | 85.8 | 81.3 | 80.8 | **88.9** | 88.0 |
+| Workspace-Bench-Lite | 70.0 | 68.2 | 62.1 | 64.9 | 68.7 | **71.7** | 60.5 | 70.1 |
+| Workspace-Bench | 66.1 | 65.0 | 58.2 | 63.5 | **68.0** | 67.4 | 56.0 | 65.8 |
+| PaperBench | 89.6 | 87.0 | 87.4 | 80.5 | **94.1** | 86.9 | 92.8 | 88.9 |
+| JobBench | 57.5 | 50.3 | 54.1 | 54.3 | 52.7 | 58.2 | 45.4 | **68.0** |
+| GDPval | 1594.2 | 1583.8 | 1493.3 | 1606.8 | 1710.9 | 1643.1 | 1611.1 | **1714.0** |
+| DeepResearch Bench II | 51.2 | 51.1 | 46.6 | 51.3 | 49.2 | 52.7 | 50.7 | **54.1** |
+| SpreadsheetBench 2 | 33.0 | 31.5 | 29.3 | **35.8** | 34.9 | 32.1 | 32.7 | 33.6 |
+| τ³-bench Banking | 43.3 | 41.2 | 44.3 | 37.1 | **55.2** | 40.2 | 46.9 | 48.7 |
+| Terminal-Bench 2.1 | 81.0 | 78.3 | 77.0 | 84.0 | 89.6 | 86.0 | 86.2 | **90.2** |
+| SWE-bench Pro | 61.7 | 59.6 | 58.4 | 62.0 | 65.3 | 60.3 | 59.8 | **76.5** |
+| NL2Repo-Bench | 50.0 | 51.4 | 49.5 | 63.1 | 63.2 | 54.8 | 60.7 | **79.9** |
+| Harbor-Index | 17.3 | 17.3 | 17.0 | 24.3 | 25.8 | 20.0 | 35.8 | **41.0** |
+| DeepSWE | 49.3 | – | 54.9 | 70.6 | 67.0 | 67.3 | 70.6 | **72.6** |
+| Harvey LAB | **27.5** | – | 6.7 | 25.4 | 24.2 | 22.5 | 5.0 | 22.7 |
+| BFCL V4 | **78.4** | 77.0 | 71.4 | 69.1 | – | 74.1 | – | – |
+| DeepSearchQA | **96.5** | 96.0 | – | 95.9 | – | 94.7 | 93.2 | – |
+| OfficeQA Pro | 67.1 | – | 66.2 | 65.4 | 66.2 | 65.8 | 66.4 | **67.2** |
+| BrowseComp | 91.3 | **92.5** | 83.4 | 91.2 | – | 89.6 | 92.2 | 90.8 |
+
+</div>
+
+> **加粗**表示该行最高分，`–` 表示结果不可用。
+
+
+<!-- ## How to Run Locally
+
+在此补充本地运行、推理服务和环境配置说明。
+
+```bash
+# 在此添加安装和运行命令。
+``` -->
+
+## 部署和在线使用
+
+Atria Dawn 支持本地部署与在线调用。有关在线调用，请使用您所在地区对应的服务。
+
+| 地区                | 使用链接                                                                  | 教程                                                                        |
+| ----------------- | --------------------------------------------------------------------- | ------------------------------------------------------------------------- |
+| **国内**         | [Link](https://discovery.intern-ai.org.cn/token-plan/home?tabIndex=0) | [Tutorial](https://discovery.intern-ai.org.cn/token-plan/home?tabIndex=3) |
+| **海外** | [Link](https://api.atria-asi.ai/)                                     | [Tutorial](https://api.atria-asi.ai/docs)                                 |
+
+如需进行本地部署，请参考以下部署指南。
+
+- [SGLang](https://github.com/sgl-project/sglang) (v0.5.13.post1+) — 参考 [cookbook](https://cookbook.sglang.io/autoregressive/GLM/GLM-5.2)
+- [vLLM](https://github.com/vllm-project/vllm) (v0.23.0+) — 参考 [recipes](https://recipes.vllm.ai/zai-org/GLM-5.2)
+
+### Codex
+
+将自定义 provider 添加到 `~/.codex/config.toml`。Codex 使用 Responses API。
+
+```toml
+model = "Atria-Dawn"
+model_provider = "atria"
+
+[model_providers.atria]
+name = "Atria"
+base_url = "https://api.atria-asi.ai/v1"
+env_key = "ATRIA_API_KEY"
+wire_api = "responses"
+```
+
+#### 将输入限制为纯文本
+
+**Atria Dawn** 只接受文本输入。Codex 默认会假设每个模型都支持多模态输入，并会附加通过 `-i/--image` 或 TUI 粘贴的图片；而该端点会因此返回 `400 Atria-Dawn is not a multimodal model` 错误。需要声明模型支持的输入模态，让 Codex 在客户端直接移除图片输入。
+
+##### 第 1 步：创建模型目录文件
+
+将下面的内容保存为 `~/.codex/atria-catalog.json`：
+
+```json
+{
+  "models": [
+    {
+      "slug": "Atria-Dawn",
+      "display_name": "Atria Dawn",
+      "base_instructions": "你是运行在 Codex CLI 中的编程代理。你将在共享工作区中与用户协作，完成软件工程目标。\n\n你只能接收文本输入。你无法获取图片、截图、PDF 或其他二进制附件。如果用户提到你无法查看的附件，请直接说明这一点，并要求用户粘贴相关文本。",
+      "supported_reasoning_levels": [
+        { "effort": "low", "description": "快速响应，使用较轻量的推理" },
+        { "effort": "medium", "description": "在速度和推理深度之间保持平衡" },
+        { "effort": "high", "description": "为复杂问题提供更深入的推理" }
+      ],
+      "shell_type": "unified_exec",
+      "visibility": "list",
+      "supported_in_api": true,
+      "priority": 1,
+      "support_verbosity": false,
+      "truncation_policy": { "mode": "tokens", "limit": 10000 },
+      "experimental_supported_tools": [],
+      "context_window": 256000,
+      "max_context_window": 256000,
+      "input_modalities": ["text"]
+    }
+  ]
+}
+```
+
+`"input_modalities": ["text"]` 是禁用多模态输入的设置。
+
+将 `context_window` / `max_context_window` 设置为模型实际支持的上限。Codex 会使用这些值来规划提示词，并判断何时自动压缩上下文。如果不设置，Codex 会采用较为保守的默认值，从而浪费可用的上下文空间。
+
+解析器要求其他所有字段都必须存在。缺少任何一个字段都会导致 `missing field <name>` 错误，Codex 也将无法启动。
+
+##### 第 2 步：让配置指向该文件
+
+```toml
+model = "Atria-Dawn"
+model_provider = "atria"
+model_catalog_json = "~/.codex/atria-catalog.json"
+
+[features]
+view_image = false
+
+[model_providers.atria]
+name = "Atria"
+base_url = "https://api.atria-asi.ai/v1"
+env_key = "ATRIA_API_KEY"
+wire_api = "responses"
+```
+
+`features.view_image = false` 是可选设置。它会移除图像查看工具，避免模型尝试调用一个会被拒绝的工具。
+
+> **重要：** `model_catalog_json` 会**替换**模型目录，而不是与其合并。任何未列在该文件中的模型都会回退到默认元数据，而默认元数据会假设模型支持多模态输入，并记录 `warning: Model metadata for <slug> not found` 警告。如果你使用 `-m` 切换模型，或通过编辑 `model` 切换模型，请将该模型也添加到同一个文件中，否则纯文本限制不会应用于该模型。
+
+要求 Codex CLI 版本为 0.154.0 或更高版本。
+
+### Claude Code
+
+```python
+#!/usr/bin/env python3
+"""
+PreToolUse 钩子：阻止 Read 工具读取 PDF 和图片文件。
+"""
+import json
+import sys
+
+
+IMAGE_EXTENSIONS = (
+    ".apng",
+    ".avif",
+    ".bmp",
+    ".gif",
+    ".heic",
+    ".heif",
+    ".ico",
+    ".jfif",
+    ".jpeg",
+    ".jpg",
+    ".jxl",
+    ".png",
+    ".svg",
+    ".tif",
+    ".tiff",
+    ".webp",
+)
+
+
+def main():
+    hook_input = json.loads(sys.stdin.read())
+    file_path = hook_input.get("tool_input", {}).get("file_path", "")
+
+    if file_path.lower().endswith(".pdf"):
+        reason = "不允许使用 Read 工具读取 PDF 文件。"
+    elif file_path.lower().endswith(IMAGE_EXTENSIONS):
+        reason = "不允许使用 Read 工具读取图片文件。"
+    else:
+        sys.exit(0)
+
+    output = {
+        "hookSpecificOutput": {
+            "hookEventName": "PreToolUse",
+            "permissionDecision": "deny",
+            "permissionDecisionReason": reason,
+        }
+    }
+    print(json.dumps(output, ensure_ascii=False))
+
+    sys.exit(0)
+
+
+if __name__ == "__main__":
+    main()
+```
+
+请将上面的代码保存为 `${Target_dir}/block_pdf_image_read.py`，并确保使用绝对路径。
+
+将它添加到 `~/claude_dir/settings.json` 中。
+
+这样可以在 `PreToolUse` 阶段拦截图片、PDF 等多模态输入。
+
+```bash
+"hooks": {
+    "PreToolUse": [
+      {
+        "matcher": "Read",
+        "hooks": [
+          {
+            "type": "command",
+            "command": "python3 ${Target_dir}/block_pdf_image_read.py",
+            "timeout": 5
+          }
+        ]
+      },
+      {
+        "matcher": "",
+        "hooks": []
+      }
+    ]
+}
+```
+### Kimi Code
+
+将自定义 provider 添加到 `~/.kimi-code/config.toml`。Kimi Code 通过兼容 OpenAI 的 Chat Completions API 与 Atria 通信。
+
+本示例适用于 Kimi Code 0.43.1 及更高版本。请先检查版本；不同版本的配置字段可能有所差异。
+
+```bash
+kimi --version
+cp -a ~/.kimi-code/config.toml ~/.kimi-code/config.toml.bak
+```
+
+#### 1. 将 provider 和模型写入用户配置
+
+将以下内容合并到 `~/.kimi-code/config.toml`。将 `default_model` 保持在顶层，并放在各个表之前。请更新已有条目，不要重复添加字段或表。将你的 `atr_...` 密钥填入 `api_key` 行中。密钥在 Atria 控制台创建，并且只会显示一次。
+
+```toml
+default_model = "atria/Atria-Dawn"
+
+[providers.atria]
+type = "openai"
+# 必须包含 /v1，Kimi 会原样追加 /chat/completions
+base_url = "https://api.atria-asi.ai/v1"
+# 只能使用字面量：不会展开 ${VAR}，也不会读取 OPENAI_API_KEY
+api_key = "ATRIA_API_KEY"
+
+[models."atria/Atria-Dawn"]
+provider = "atria"
+# 区分大小写的实际请求名称，与上面的表名不同（上面的表名是本地别名）
+model = "Atria-Dawn"
+display_name = "Atria Dawn"
+# 必填：缺少此项时，该条目会静默加载失败。请��持为实际限制值，
+# 这样上下文预算和压缩功能才能正常工作。
+max_context_size = 256000
+# 发送到接口的 max_tokens 值。缺少此项时，Kimi 会改为发送 max_context_size，
+# Atria 会拒绝该请求（有效范围为 1-65536）。
+max_output_size = 65536
+# 这是纯文本模型：添加 image_in 会失败，并提示 "is not a multimodal model"。
+# 请明确声明 tool_use，否则 Kimi 会将功能视为未知。
+# "thinking" 用于开启推理功能；如果省略，effort 会被强制设为
+# "off"，并且不会发送 reasoning_effort。
+capabilities = ["tool_use", "thinking"]
+# 如果没有 support_efforts，effort 会保持为不透明的 "on"，完全不会发送
+# reasoning_effort。列出这些选项后，/think 命令和选择器才能正常工作。
+support_efforts = ["low", "medium", "high", "xhigh", "max"]
+default_effort = "max"
+# 必填，用于允许关闭 thinking。省略此项会失败，并提示 "reasons by
+# default but declares no off effort"。
+off_effort = "none"
+```
+
+由于密钥以明文存储，请确保文件权限保持私密：
+
+```bash
+chmod 600 ~/.kimi-code/config.toml
+```
+
+#### 2. 重启并验证
+
+验证配置，然后发送一个纯文本任务。
+
+```bash
+kimi doctor
+kimi provider list
+kimi
+```
+
+预期输出：
+
+```
+OK config.toml  /home/<user>/.kimi-code/config.toml
+All checked config files are valid.
+
+atria  type=openai  models=1  source=inline
+Default model: atria/Atria-Dawn
+```
+
+如果 provider 列表中没有 `atria`，说明模型条目解析失败，最常见的原因是缺少 `max_context_size`。
+
+
+## 许可证
+
+本仓库的代码和权重依照MIT协议开源。
+
+<!-- ## Citation
+
+```text
+在此添加正式引用信息。
+``` -->
+
+## 联系我们
+如有问题或建议，欢迎通过邮件或GitHub等平台联系我们。
+
+- 📱 **微信社群** — 扫描下方二维码加入 Atria 交流群⑦
+
+<p align="center">
+  <img src="assets/wechat-community-7.png" alt="Atria 微信交流群 7 二维码" width="320" />
+</p>
+
+## 引用
+```
+@misc{guo2026atriadawndawnagentic,
+      title={Atria Dawn: The Dawn of Agentic Superintelligence}, 
+      author={Honglin Guo and Tao Gui and Yicheng Chen and Guanting Dong and Qiming Ge and Yuyang Hu and Zixian Huang and Jiajie Jin and Alexander Lam and Yining Li and Jiahang Lin and Yanjiang Liu and Xinyu Lu and Haijun Lv and Junlin Shang and Qisheng Su and Guoqiang Wang and Rui Wang and Zhecan Wang and Hao Xiang and Xinchen Xie and Shuhao Xing and Xiaoyu Xing and Wanghan Xu and Xinyu Yang and Yajie Yang and Chengfeng Zhao and Haoran Zhao and Ruojun Zhou and Yunhua Zhou and Yicheng Zou and Kun Cai and Qiye Cai and Xinmeng Che and Haodong Chen and Jiabei Chen and Jiahao Chen and Jiayi Chen and Yujia Chen and Lizhi Cui and Youheng Dai and Xin Deng and Yi Dong and Shihan Dou and Chenya Gu and Xu Guo and Ding Han and Feiyang Hao and Haotan He and Jie Hou and Binze Hu and Zijian Hu and Junhao Huang and Huicheng Jiang and Jiazhen Jiang and Shufan Jiang and Jiahao Kuang and Bowen Lai and Bo Li and Jiaqiang Li and Peng Li and Qilong Li and Zhuoqun Li and Jiaxiang Liu and Shuainan Liu and Tong Liu and Yi Liu and Zhonghang Lu and Jianwen Luo and Yanyi Luo and Huijie Lv and Ningsheng Ma and Zerun Ma and Houcheng Min and Chengjun Pan and Qiyuan Peng and Xiaoxuan Peng and Jianmin Qian and Jiantao Qiu and Wanying Ren and Huayu Sha and Jifei Shan and Zixin Shang and Bing Shao and Zhuohui Sheng and Jiayang Shi and Yang Shu and Aierpanjiang Simayi and Sirui Song and Yuxiao Song and Zhe Sun and Zhichao Sun and Wenzhe Tan and Wenhui Tian and Zhongbo Tian and Hanchen Wang and Pengbo Wang and Rui Wang and Yiding Wang and Yuhui Wang and Zhiheng Xi and Caijun Xu and Chao Xu and Yongfeng Xu and Xiaolei Yang and Zhixiong Yang and Qian Yao and Shihong Yi and Yuankai Ying and Jia Yu and Dingbo Yuan and Hao Yuan and Junjie Yuan and Bo Zhang and Caixian Zhang and Qiuyinzhe Zhang and Jiyuan Zhao and Penghao Zhao and Ying Zhao and Pujun Zheng and Xiaoxue Zhong and Xiaohao Zhou and Xinyu Zhou and Dongsheng Zhu and Guanru Zhu and Yulun Zhu and Yaojie Lu and Tao Ji and Hongyu Lin and Yutao Zhu and Pengfei Cao and Guoxiu He and Xianpei Han and Ben He and Zhicheng Dou and Kang Liu and Qi Zhang and Le Sun and Jun Zhao and Ji-Rong Wen and Xuanjing Huang and Yu-Gang Jiang and Bowen Zhou},
+      year={2026},
+      eprint={2609.15818},
+      archivePrefix={arXiv},
+      primaryClass={cs.AI},
+      url={https://arxiv.org/abs/2609.15818}, 
+}
+```