Publish LocateAnything-3B RKNN artifacts + worker manifest

.gitattributes

@@ -1,3 +1,12 @@
+# 本项目模型分发的 LFS 规则。
+#
+# 前半段 = Hub 新建仓的系统默认模板（保持一致，避免丢格式）；
+# 后半段 = 本仓库的权重，**按完整路径逐个点名**。
+#
+# 注意：worker manifest（<model_id>.json）必须保持**纯文本**
+#（要能 diff / 审阅 / 网页预览），所以这里绝不能对 json 扩展名
+# 做通配兜底。
+
 *.7z filter=lfs diff=lfs merge=lfs -text
 *.arrow filter=lfs diff=lfs merge=lfs -text
 *.bin filter=lfs diff=lfs merge=lfs -text
@@ -33,3 +42,11 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+
+# ---- 本仓库权重 ----
+LocateAnything.embed.bin filter=lfs diff=lfs merge=lfs -text
+LocateAnything.tokenizer.gguf filter=lfs diff=lfs merge=lfs -text
+llm_LocateAnything.1828.rknn filter=lfs diff=lfs merge=lfs -text
+llm_LocateAnything.weight filter=lfs diff=lfs merge=lfs -text
+vision_LocateAnything.448x448.1828.rknn filter=lfs diff=lfs merge=lfs -text
+vision_LocateAnything.448x448.weight filter=lfs diff=lfs merge=lfs -text

LICENSE

@@ -0,0 +1,74 @@
+                                   NVIDIA License
+
+1. Definitions
+
+“Licensor” means any person or entity that distributes its Work.
+
+“Work” means (a) the original work of authorship made available under this license, 
+which may include software, documentation, or other files, and (b) any additions to or 
+derivative works  thereof  that are made available under this license.
+
+The terms “reproduce,” “reproduction,” “derivative works,” and “distribution” have the 
+meaning as provided under U.S. copyright law; provided, however, that for the purposes 
+of this license, derivative works shall not include works that remain separable from, or 
+merely link (or bind by name) to the interfaces of, the Work.
+
+Works are “made available” under this license by including in or with the Work either 
+(a) a copyright notice referencing the applicability of this license to the Work, or 
+(b) a copy of this license.
+
+2. License Grant
+
+2.1 Copyright Grant. Subject to the terms and conditions of this license, each Licensor 
+grants to you a perpetual, worldwide, non-exclusive, royalty-free, copyright license to 
+use, reproduce, prepare derivative works of, publicly display, publicly perform, 
+sublicense and distribute its Work and any resulting derivative works in any form.
+
+3. Limitations
+
+3.1 Redistribution. You may reproduce or distribute the Work only if (a) you do so under 
+this license, (b) you include a complete copy of this license with your distribution, 
+and (c) you retain without modification any copyright, patent, trademark, or attribution 
+notices that are present in the Work.
+
+3.2 Derivative Works. You may specify that additional or different terms apply to the 
+use, reproduction, and distribution of your derivative works of the Work (“Your Terms”) 
+only if (a) Your Terms provide that the use limitation in Section 3.3 applies to your 
+derivative works, and (b) you identify the specific derivative works that are subject 
+to Your Terms. Notwithstanding Your Terms, this license (including the redistribution 
+requirements in Section 3.1) will continue to apply to the Work itself.
+
+3.3 Use Limitation. The Work and any derivative works thereof only may be used or 
+intended for use non-commercially. Notwithstanding the foregoing, NVIDIA Corporation 
+and its affiliates may use the Work and any derivative works commercially. As used 
+herein, “non-commercially” means for research or evaluation purposes only.
+
+3.4 Patent Claims. If you bring or threaten to bring a patent claim against any 
+Licensor (including any claim, cross-claim or counterclaim in a lawsuit) to enforce any 
+patents that you allege are infringed by any Work, then your rights under this license 
+from such Licensor (including the grant in Section 2.1) will terminate immediately.
+
+3.5 Trademarks. This license does not grant any rights to use any Licensor’s or its 
+affiliates’ names, logos, or trademarks, except as necessary to reproduce the notices 
+described in this license.
+
+3.6 Termination. If you violate any term of this license, then your rights under this 
+license (including the grant in Section 2.1) will terminate immediately.
+
+4. Disclaimer of Warranty.
+
+THE WORK IS PROVIDED “AS IS” WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, EITHER 
+EXPRESS OR IMPLIED, INCLUDING WARRANTIES OR CONDITIONS OF MERCHANTABILITY, FITNESS FOR 
+A PARTICULAR PURPOSE, TITLE OR NON-INFRINGEMENT. YOU BEAR THE RISK OF UNDERTAKING ANY 
+ACTIVITIES UNDER THIS LICENSE. 
+
+5. Limitation of Liability.
+
+EXCEPT AS PROHIBITED BY APPLICABLE LAW, IN NO EVENT AND UNDER NO LEGAL THEORY, WHETHER 
+IN TORT (INCLUDING NEGLIGENCE), CONTRACT, OR OTHERWISE SHALL ANY LICENSOR BE LIABLE TO 
+YOU FOR DAMAGES, INCLUDING ANY DIRECT, INDIRECT, SPECIAL, INCIDENTAL, OR CONSEQUENTIAL 
+DAMAGES ARISING OUT OF OR RELATED TO THIS LICENSE, THE USE OR INABILITY TO USE THE WORK 
+(INCLUDING BUT NOT LIMITED TO LOSS OF GOODWILL, BUSINESS INTERRUPTION, LOST PROFITS OR 
+DATA, COMPUTER FAILURE OR MALFUNCTION, OR ANY OTHER DAMAGES OR LOSSES), EVEN IF THE 
+LICENSOR HAS BEEN ADVISED OF THE POSSIBILITY OF SUCH DAMAGES.
+

LocateAnything-3b.json

@@ -0,0 +1,19 @@
+{
+  "label": "[GD]视觉定位模型",
+  "capability_types": [
+    "swcap_vision_grounding"
+  ],
+  "spec_factory": "locate_anything",
+  "worker_name": "ground",
+  "model_dir_name": "LocateAnything-3B",
+  "model_dir_env": "WDNAI_LOCATE_ANYTHING_DIR",
+  "instance_name": "llm-ground-3b",
+  "memory_estimate_mb": 5500,
+  "estimated_init_seconds": 60,
+  "description": "视觉定位模型，给定文本描述输出图像中目标区域",
+  "notes": "视觉定位（kind=\"grounding\"，0-1000 归一化 bbox）。权重目录由 wdnai 内部的 _default_locate_anything_paths 解析，可用 WDNAI_LOCATE_ANYTHING_DIR 覆写。模型未部署时 load 会抛 FileNotFoundError，由 worker 的 load 流程转成 errored 状态。",
+  "available_targets": [
+    "182x"
+  ],
+  "display_order": 70
+}

LocateAnything.embed.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ce1515f14bc5dec19664994e105e22f19d0c682c55a19b04260a15be17178654
+size 625381376

LocateAnything.tokenizer.gguf

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ca08804b1187ed778fd02ff9039cc3c7d2147d2e54723b3fe53b5f64e513e219
+size 5942707

NOTICE

@@ -0,0 +1,32 @@
+Source model: nvidia/LocateAnything-3B
+Source URL: https://huggingface.co/nvidia/LocateAnything-3B
+Source artifact(s) and SHA-256:
+  assets/decoding_demo.mp4
+    SHA-256: 3f4083674bacb26f43d1f68f5529dac7ed4c713dca724d90b718358253453aba
+  assets/demo.mp4
+    SHA-256: 56e5c216d26741be466562e90bf6f3d576f6e3c7ac64341a9101cccf94af2335
+  model-00001-of-00002.safetensors
+    SHA-256: 923cfc10fed19808067da6df85a9a4220ddc1f9eb91ceee94c0fecd05d0f2d58
+  model-00002-of-00002.safetensors
+    SHA-256: 3459ba101f40594f3f62d3312014f1f8378b4ba3da3b1d562480045938fc7d47
+Source revision: not recoverable from conversion records
+Source license: NVIDIA License
+
+The files in this repository are RKNN-converted derivatives, not unmodified upstream weights.
+
+Artifact names and formats differ from upstream on purpose: the upstream checkpoint is a
+single safetensors file per stage and is split here into RKNN graphs plus their
+companion .weight files, and the upstream tokenizer.json was converted into a GGUF
+tokenizer. The SHA-256 values above identify the **upstream** artifacts, not the files
+in this repository — use SHA256SUMS for the latter.
+
+Files added by this repository (not part of the upstream model)
+--------------------------------------------------------------
+  LocateAnything-3b.json  wdnwebui worker manifest — the "manual" of this distribution unit.
+                  Deliberately plain text so it can be diffed, reviewed and
+                  previewed on the Hub.
+  config.json     Release metadata and file manifest. This is NOT the upstream
+                  architecture config.json.
+  SHA256SUMS      SHA-256 checksums of every model file in this repository.
+  README.md,
+  README_CN.md    Documentation.

README.md

@@ -0,0 +1,116 @@
+---
+license: other
+library_name: rkllm
+tags:
+- rk182x
+- rockchip
+- rknpu
+- vlm
+- qwen
+- edge-ai
+---
+
+# LocateAnything-3B — [GD]视觉定位模型
+
+[English](README.md) | [简体中文](README_CN.md)
+
+## 1. Overview
+
+视觉定位模型，给定文本描述输出图像中目标区域
+
+This repository ships the RK182x deployment artifacts (RKNN graphs +
+weights) **together with the wdnwebui worker manifest**
+(`LocateAnything-3b.json`).
+
+| Item | Value |
+| --- | --- |
+| Model ID | `wardner/LocateAnything-3B` |
+| Display name | LocateAnything-3B |
+| Model type | VLM |
+| Compute target | RK182x |
+| Host platform | RK3588 |
+| Runtime | RKLLM + RKNN |
+| Capabilities | `swcap_vision_grounding` |
+| Worker factory | `locate_anything` |
+| Memory estimate | ~5500 MB |
+| Init time | ~60 s |
+| Cards required | 1 |
+| Total weight size | 3.27 GB |
+
+## 2. Files
+
+| Path | Description | Size |
+| --- | --- | --- |
+| `LocateAnything.embed.bin` | fp16 embedding table (vocab × hidden) | 596.4 MB |
+| `LocateAnything.tokenizer.gguf` | Tokenizer (GGUF, GPT-2 BPE + ByteLevel) | 5.7 MB |
+| `llm_LocateAnything.1828.rknn` | RKNN graph (RK182x compute card, 1.82 TOPS generation) | 43.0 MB |
+| `llm_LocateAnything.weight` | Weights paired with the RKNN graph above | 1848.4 MB |
+| `vision_LocateAnything.448x448.1828.rknn` | RKNN graph (RK182x compute card, 1.82 TOPS generation) | 8.1 MB |
+| `vision_LocateAnything.448x448.weight` | Weights paired with the RKNN graph above | 847.5 MB |
+| `LocateAnything-3b.json` | **worker manifest** — deliberately plain text | 793 B |
+
+> `LocateAnything-3b.json` is **deliberately not** in Git LFS. It is the
+> manual of this distribution unit and must stay plain text so it can be
+> diffed, reviewed and previewed on the Hub.
+
+## 3. Download
+
+```bash
+hf download wardner/LocateAnything-3B --local-dir ./LocateAnything-3B
+```
+
+Or with git (requires Git LFS):
+
+```bash
+git clone https://huggingface.co/wardner/LocateAnything-3B
+```
+
+If `huggingface.co` is unreachable from your network (DNS interception /
+blocked access), point the client at a mirror instead:
+
+```bash
+HF_ENDPOINT=https://hf-mirror.com hf download wardner/LocateAnything-3B \
+  --local-dir ./LocateAnything-3B
+```
+
+Note: with `git clone` the weights are Git LFS **pointers** until
+`git lfs pull` completes. `hf download` fetches the real files directly, and
+does not create a `.git` directory — neither is required for the model to
+work; see §4.
+
+## 4. Installation (wdnwebui)
+
+Clone the repository into the model root so the folder holds **both** the
+weights and the worker manifest:
+
+```bash
+git clone https://huggingface.co/wardner/LocateAnything-3B /userdata/models/LocateAnything-3B
+```
+
+Then open the web UI → **Compute** → the card appears under "local compute"
+(the backend rescans the model root every ~2 s, **no restart needed**) →
+click "enable".
+
+The directory name does not have to match the model name: the SDK follows
+"the folder the manifest was found in", located through the
+`WDNAI_LOCATE_ANYTHING_DIR` environment variable declared in the manifest.
+
+## 5. SHA-256 verification
+
+```bash
+cd ./LocateAnything-3B
+sha256sum -c SHA256SUMS
+```
+
+All entries must report `OK` before deployment.
+
+## 6. Caveats and limits
+
+See the `notes` field of `LocateAnything-3b.json` — it records the
+per-model specifics (sampling defaults, capability layout, side effects).
+
+## 7. License
+
+The upstream model is released under the Apache License 2.0. The files here
+are RKNN-converted derivatives, not unmodified upstream weights. See
+[LICENSE](LICENSE) and [NOTICE](NOTICE).

README_CN.md

@@ -0,0 +1,109 @@
+---
+license: other
+library_name: rkllm
+tags:
+- rk182x
+- rockchip
+- rknpu
+- vlm
+- qwen
+- edge-ai
+---
+
+# LocateAnything-3B —— [GD]视觉定位模型
+
+[English](README.md) | [简体中文](README_CN.md)
+
+## 1. 模型简介
+
+视觉定位模型，给定文本描述输出图像中目标区域
+
+本仓库提供 RK182x 协卡上的部署产物（RKNN 图 + 权重），**并且带上
+wdnwebui 算力层需要的 worker 配置**（`LocateAnything-3b.json`）。
+
+| 项 | 值 |
+| --- | --- |
+| 模型 ID | `wardner/LocateAnything-3B` |
+| 显示名 | LocateAnything-3B |
+| 模型类型 | VLM |
+| 目标协卡 | RK182x |
+| 宿主平台 | RK3588 |
+| 运行时 | RKLLM + RKNN |
+| 能力 | `swcap_vision_grounding` |
+| worker 工厂 | `locate_anything` |
+| 显存预估 | 约 5500 MB |
+| 初始化耗时 | 约 60 秒 |
+| 占用卡数 | 1 |
+| 权重总大小 | 3.27 GB |
+
+## 2. 文件说明
+
+| 路径 | 说明 | 大小 |
+| --- | --- | --- |
+| `LocateAnything.embed.bin` | fp16 embedding 表（vocab × hidden） | 596.4 MB |
+| `LocateAnything.tokenizer.gguf` | 分词器（GGUF，GPT-2 BPE + ByteLevel） | 5.7 MB |
+| `llm_LocateAnything.1828.rknn` | RKNN 图（RK182x 协卡，1.82 TOPS 代） | 43.0 MB |
+| `llm_LocateAnything.weight` | 与上面 RKNN 图配套的权重 | 1848.4 MB |
+| `vision_LocateAnything.448x448.1828.rknn` | RKNN 图（RK182x 协卡，1.82 TOPS 代） | 8.1 MB |
+| `vision_LocateAnything.448x448.weight` | 与上面 RKNN 图配套的权重 | 847.5 MB |
+| `LocateAnything-3b.json` | **worker 配置** —— 刻意保持纯文本 | 793 B |
+
+> `LocateAnything-3b.json` **刻意不**放进 Git LFS。它是这个分发单元的
+> 「说明书」，必须保持纯文本，才能被 diff、被审阅、也能在 Hub 网页上直接预览。
+
+## 3. 下载
+
+```bash
+hf download wardner/LocateAnything-3B --local-dir ./LocateAnything-3B
+```
+
+或者用 git（需要安装 Git LFS）：
+
+```bash
+git clone https://huggingface.co/wardner/LocateAnything-3B
+```
+
+如果你的网络访问不了 `huggingface.co`（DNS 污染 / 访问被阻断），改用镜像：
+
+```bash
+HF_ENDPOINT=https://hf-mirror.com hf download wardner/LocateAnything-3B \
+  --local-dir ./LocateAnything-3B
+```
+
+注意：用 `git clone` 时，权重在 `git lfs pull` 完成前只是 **LFS 指针**；
+而 `hf download` 直接下载真实文件，**不会生成 `.git` 目录**。两种方式都
+不需要 `.git` —— 见 §4。
+
+## 4. 安装（wdnwebui）
+
+把仓库克隆到模型根目录，让这个文件夹**同时**包含权重和 worker 配置：
+
+```bash
+git clone https://huggingface.co/wardner/LocateAnything-3B /userdata/models/LocateAnything-3B
+```
+
+然后打开网页 → **算力配置** → 卡片会出现在「本地算力」里（后端每 ~2 秒
+重扫一次模型根目录，**不需要重启服务**）→ 点「启用」即可。
+
+目录名不必和模型名一致：SDK 跟随的是「manifest 所在的那个文件夹」，
+通过 manifest 里声明的 `WDNAI_LOCATE_ANYTHING_DIR` 环境变量定位。
+
+## 5. SHA-256 校验
+
+```bash
+cd ./LocateAnything-3B
+sha256sum -c SHA256SUMS
+```
+
+部署前所有条目都必须报 `OK`。
+
+## 6. 注意事项与限制
+
+- 视觉定位（kind="grounding"，0-1000 归一化 bbox）
+- 权重目录由 wdnai 内部的 _default_locate_anything_paths 解析，可用 WDNAI_LOCATE_ANYTHING_DIR 覆写
+- 模型未部署时 load 会抛 FileNotFoundError，由 worker 的 load 流程转成 errored 状态。
+
+## 7. 版权与许可
+
+上游模型以 Apache License 2.0 发布。本仓库中的文件是 RKNN 转换后的衍生
+产物，并非未经修改的上游权重。详见 [LICENSE](LICENSE) 与 [NOTICE](NOTICE)。

SHA256SUMS

@@ -0,0 +1,7 @@
+7670ab40a6d5898ee69e7e1556a87b65b178507e4211a3085f396446d055c7a2  LocateAnything-3b.json
+ce1515f14bc5dec19664994e105e22f19d0c682c55a19b04260a15be17178654  LocateAnything.embed.bin
+ca08804b1187ed778fd02ff9039cc3c7d2147d2e54723b3fe53b5f64e513e219  LocateAnything.tokenizer.gguf
+06141c21ace8465c8e49cbeb84370f875d5cefe592ba312e39dab617a263d5cd  llm_LocateAnything.1828.rknn
+37b4ec3ca6725fed3f5c755a5248dc4b7c4fe8c84e9c37c7ea05e95185b4b2cd  llm_LocateAnything.weight
+777f2cf4dd7245f85f009d8bb375962294f274ec985aba75454e3162cf77140e  vision_LocateAnything.448x448.1828.rknn
+5e930ed7e27ccb361063fd51c56039b99047f29ae83942e8f793b3c3962858b4  vision_LocateAnything.448x448.weight

config.json

@@ -0,0 +1,70 @@
+{
+  "schema_version": "1.0",
+  "model_id": "wardner/LocateAnything-3B",
+  "display_name": "LocateAnything-3B",
+  "model_type": "VLM",
+  "chip_label": "RK182x",
+  "supported_chips": [
+    "RK182x"
+  ],
+  "host_platform": "RK3588",
+  "runtime_type": "rkllm+rknn",
+  "worker": {
+    "id": "locate_anything",
+    "label": "[GD]视觉定位模型",
+    "capability_types": [
+      "swcap_vision_grounding"
+    ],
+    "model_dir_env": "WDNAI_LOCATE_ANYTHING_DIR",
+    "model_dir_name": "LocateAnything-3B",
+    "available_targets": [
+      "182x"
+    ],
+    "memory_estimate_mb": 5500,
+    "estimated_init_seconds": 60
+  },
+  "files": [
+    {
+      "path": "LocateAnything.embed.bin",
+      "size": 625381376,
+      "lfs": true,
+      "sha256": "ce1515f14bc5dec19664994e105e22f19d0c682c55a19b04260a15be17178654"
+    },
+    {
+      "path": "LocateAnything.tokenizer.gguf",
+      "size": 5942707,
+      "lfs": true,
+      "sha256": "ca08804b1187ed778fd02ff9039cc3c7d2147d2e54723b3fe53b5f64e513e219"
+    },
+    {
+      "path": "llm_LocateAnything.1828.rknn",
+      "size": 45101424,
+      "lfs": true,
+      "sha256": "06141c21ace8465c8e49cbeb84370f875d5cefe592ba312e39dab617a263d5cd"
+    },
+    {
+      "path": "llm_LocateAnything.weight",
+      "size": 1938139648,
+      "lfs": true,
+      "sha256": "37b4ec3ca6725fed3f5c755a5248dc4b7c4fe8c84e9c37c7ea05e95185b4b2cd"
+    },
+    {
+      "path": "vision_LocateAnything.448x448.1828.rknn",
+      "size": 8544688,
+      "lfs": true,
+      "sha256": "777f2cf4dd7245f85f009d8bb375962294f274ec985aba75454e3162cf77140e"
+    },
+    {
+      "path": "vision_LocateAnything.448x448.weight",
+      "size": 888622592,
+      "lfs": true,
+      "sha256": "5e930ed7e27ccb361063fd51c56039b99047f29ae83942e8f793b3c3962858b4"
+    },
+    {
+      "path": "LocateAnything-3b.json",
+      "size": 793,
+      "lfs": false,
+      "sha256": "7670ab40a6d5898ee69e7e1556a87b65b178507e4211a3085f396446d055c7a2"
+    }
+  ]
+}

llm_LocateAnything.1828.rknn

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:06141c21ace8465c8e49cbeb84370f875d5cefe592ba312e39dab617a263d5cd
+size 45101424

llm_LocateAnything.weight

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:37b4ec3ca6725fed3f5c755a5248dc4b7c4fe8c84e9c37c7ea05e95185b4b2cd
+size 1938139648

vision_LocateAnything.448x448.1828.rknn

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:777f2cf4dd7245f85f009d8bb375962294f274ec985aba75454e3162cf77140e
+size 8544688

vision_LocateAnything.448x448.weight

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5e930ed7e27ccb361063fd51c56039b99047f29ae83942e8f793b3c3962858b4
+size 888622592