Update canonical model series and public download statistics

README.md

@@ -141,6 +141,35 @@ Compute depth and output width address two different costs. The native low-effor
 
 The optimization reduces output-head work, not the context-processing backbone. Current vLLM latency results do not include candidate-only output-head execution. See the [Technical Report](TECHNICAL_REPORT.md#6-performance-where-the-savings-can-come-from) for the mechanism and execution boundaries.
 
+<!-- OPENJEV-SERIES:START -->
+## Model series and public downloads
+
+[APUS-OpenJev-v1 Collection](https://huggingface.co/collections/apus-ailab/apus-openjev-v1-6ab1ee888eb002fcdd3a2825)
+
+Standalone repositories use standard root-level model files. This repository and its legacy download paths remain available.
+
+- **4B · checkpoint-5949 · BF16**: [https://huggingface.co/apus-ailab/APUS-OpenJev-v1-4B](https://huggingface.co/apus-ailab/APUS-OpenJev-v1-4B)
+
+- **9B · checkpoint-3000 · BF16**: [https://huggingface.co/apus-ailab/APUS-OpenJev-v1-9B](https://huggingface.co/apus-ailab/APUS-OpenJev-v1-9B)
+
+- **35B-A3B · checkpoint-5949**: Merge/reload validation in progress; link available after publication.
+
+
+### Download statistics · 项目汇总
+
+Status: **current**. Observed: 2026-09-22T03:10:37.132203Z (UTC). Last checked: 2026-09-22T03:10:37.132203Z (UTC).
+
+| Model | Last 30 days | All time | Status |
+| --- | ---: | ---: | --- |
+| [4B](https://huggingface.co/apus-ailab/APUS-OpenJev-v1-4B) | 0 | 0 | current |
+| [9B](https://huggingface.co/apus-ailab/APUS-OpenJev-v1-9B) | 0 | 0 | current |
+| 35B-A3B | — | — | unpublished; excluded |
+| **Series total (project-computed)** | 0 | 0 | complete |
+
+Totals sum only the published canonical repositories in this snapshot, separately for each time window. Historical snapshots and family/mirror repositories are never added. This is a project-computed summary, not a native Hugging Face Collection counter. Hub download counters are not unique users.
+
+<!-- OPENJEV-SERIES:END -->
+
 ## 5. Minimal Inference
 
 The root [meta.yaml](meta.yaml) catalogs model variants and their configuration hashes. Include it when downloading a variant; the repository and existing model paths are unchanged. You can also use [download_model.py](download_model.py) with `--variant 9B-3000 --local-dir ./APUS-OpenJev-v1`; it selects and checks the variant at one fixed Hub revision. Hub download statistics count requests to selected metadata files, not unique users or completed weight transfers. Older subfolder-only clients can bypass this metadata.
@@ -245,30 +274,3 @@ Visit the [APUS official website](https://www.apusai.com) or [APUS AI Lab on Hug
 
 **Authors:** gumpcheng ([https://huggingface.co/xDAN2099](https://huggingface.co/xDAN2099)), zhangxu, [APUS AI-LAB](https://github.com/APUS-AI-Lab).
 
-<!-- OPENJEV-SERIES:START -->
-## Model series and public downloads
-
-[APUS-OpenJev-v1 Collection](https://huggingface.co/collections/apus-ailab/apus-openjev-v1-6ab1ee888eb002fcdd3a2825)
-
-Standalone repositories use standard root-level model files. This repository and its legacy download paths remain available.
-
-| Model | Checkpoint | Release |
-|---|---:|---|
-| [4B](https://huggingface.co/apus-ailab/APUS-OpenJev-v1-4B) | 5949 | BF16 standalone |
-| [9B](https://huggingface.co/apus-ailab/APUS-OpenJev-v1-9B) | 3000 | BF16 standalone |
-| 35B-A3B | 5949 | Merge/reload validation in progress |
-
-### Download statistics · 项目汇总
-
-Status: **current**. Observed: 2026-09-22T03:08:12.998307Z (UTC). Last checked: 2026-09-22T03:08:12.998307Z (UTC).
-
-| Model | Last 30 days | All time | Status |
-| --- | ---: | ---: | --- |
-| [4B](https://huggingface.co/apus-ailab/APUS-OpenJev-v1-4B) | 0 | 0 | current |
-| [9B](https://huggingface.co/apus-ailab/APUS-OpenJev-v1-9B) | 0 | 0 | current |
-| 35B-A3B | — | — | unpublished; excluded |
-| **Series total (project-computed)** | 0 | 0 | complete |
-
-Totals sum only the published canonical repositories in this snapshot, separately for each time window. Historical snapshots and family/mirror repositories are never added. This is a project-computed summary, not a native Hugging Face Collection counter. Hub download counters are not unique users.
-
-<!-- OPENJEV-SERIES:END -->

README.zh-CN.md

@@ -141,6 +141,35 @@ tags:
 
 这一优化减少的是输出头工作，不省略上下文骨干计算；当前 vLLM 延迟结果尚不包含候选专用输出头的收益。机制和执行范围见[技术报告](TECHNICAL_REPORT.md#6-performance-where-the-savings-can-come-from)。
 
+<!-- OPENJEV-SERIES:START -->
+## 模型系列与公开下载统计
+
+[APUS-OpenJev-v1 Collection](https://huggingface.co/collections/apus-ailab/apus-openjev-v1-6ab1ee888eb002fcdd3a2825)
+
+独立模型仓库已采用标准根目录布局；原仓库与旧下载路径继续保留。
+
+- **4B · checkpoint-5949 · BF16**: [https://huggingface.co/apus-ailab/APUS-OpenJev-v1-4B](https://huggingface.co/apus-ailab/APUS-OpenJev-v1-4B)
+
+- **9B · checkpoint-3000 · BF16**: [https://huggingface.co/apus-ailab/APUS-OpenJev-v1-9B](https://huggingface.co/apus-ailab/APUS-OpenJev-v1-9B)
+
+- **35B-A3B · checkpoint-5949**: 合并与独立加载验证中，发布后提供链接.
+
+
+### Download statistics · 项目汇总
+
+Status: **current**. Observed: 2026-09-22T03:10:37.132203Z (UTC). Last checked: 2026-09-22T03:10:37.132203Z (UTC).
+
+| Model | Last 30 days | All time | Status |
+| --- | ---: | ---: | --- |
+| [4B](https://huggingface.co/apus-ailab/APUS-OpenJev-v1-4B) | 0 | 0 | current |
+| [9B](https://huggingface.co/apus-ailab/APUS-OpenJev-v1-9B) | 0 | 0 | current |
+| 35B-A3B | — | — | unpublished; excluded |
+| **Series total (project-computed)** | 0 | 0 | complete |
+
+Totals sum only the published canonical repositories in this snapshot, separately for each time window. Historical snapshots and family/mirror repositories are never added. This is a project-computed summary, not a native Hugging Face Collection counter. Hub download counters are not unique users.
+
+<!-- OPENJEV-SERIES:END -->
+
 ## 5. 最小推理示例
 
 根目录 [meta.yaml](meta.yaml) 提供模型版本及配置哈希；下载指定版本时请同时保留它，仓库与原模型路径不变。也可运行 [download_model.py](download_model.py)，参数为 `--variant 9B-3000 --local-dir ./APUS-OpenJev-v1`，在同一固定 Hub revision 选择并校验模型。Hub 下载统计按指定元数据文件的请求计数，不等同独立用户或完整权重下载；旧版仅下载子目录的客户端可能绕过此元数据。
@@ -200,30 +229,3 @@ python examples.py . --device cuda:0 --effort high
 
 **Authors:** gumpcheng ([https://huggingface.co/xDAN2099](https://huggingface.co/xDAN2099)), zhangxu, [APUS AI-LAB](https://github.com/APUS-AI-Lab)。
 
-<!-- OPENJEV-SERIES:START -->
-## 模型系列与公开下载统计
-
-[APUS-OpenJev-v1 Collection](https://huggingface.co/collections/apus-ailab/apus-openjev-v1-6ab1ee888eb002fcdd3a2825)
-
-独立模型仓库已采用标准根目录布局；原仓库与旧下载路径继续保留。
-
-| Model | Checkpoint | Release |
-|---|---:|---|
-| [4B](https://huggingface.co/apus-ailab/APUS-OpenJev-v1-4B) | 5949 | BF16 standalone |
-| [9B](https://huggingface.co/apus-ailab/APUS-OpenJev-v1-9B) | 3000 | BF16 standalone |
-| 35B-A3B | 5949 | Merge/reload validation in progress |
-
-### Download statistics · 项目汇总
-
-Status: **current**. Observed: 2026-09-22T03:08:12.998307Z (UTC). Last checked: 2026-09-22T03:08:12.998307Z (UTC).
-
-| Model | Last 30 days | All time | Status |
-| --- | ---: | ---: | --- |
-| [4B](https://huggingface.co/apus-ailab/APUS-OpenJev-v1-4B) | 0 | 0 | current |
-| [9B](https://huggingface.co/apus-ailab/APUS-OpenJev-v1-9B) | 0 | 0 | current |
-| 35B-A3B | — | — | unpublished; excluded |
-| **Series total (project-computed)** | 0 | 0 | complete |
-
-Totals sum only the published canonical repositories in this snapshot, separately for each time window. Historical snapshots and family/mirror repositories are never added. This is a project-computed summary, not a native Hugging Face Collection counter. Hub download counters are not unique users.
-
-<!-- OPENJEV-SERIES:END -->

bundle-manifest.json

@@ -592,13 +592,13 @@
     },
     {
       "path": "README.md",
-      "bytes": 20765,
-      "sha256": "f91858cadb5712953e1dda32ea0162c97638bfd090e16858dc98398955e12e03"
+      "bytes": 20885,
+      "sha256": "2b7ddc3aad1a9a50feb09690e73d176638c9f37267696906aad15502b7006d59"
     },
     {
       "path": "README.zh-CN.md",
-      "bytes": 15931,
-      "sha256": "a0b4f4ff878d1c84a9c8a04c1464fa06776cce86221460e427621208bb45a76b"
+      "bytes": 16036,
+      "sha256": "fdac890892b1e7b80c013e56d3aa75406cba2512b5071268531732d7da7b4e9a"
     },
     {
       "path": "TECHNICAL_REPORT.md",
@@ -718,7 +718,7 @@
     {
       "path": "series-downloads.json",
       "bytes": 742,
-      "sha256": "1743e657ffcc6b8794d4852db7197fdc00bce22c9171956e463596151ede61f6"
+      "sha256": "f307b3f37691a1ab2624b5769bb32b605ed6893177d73a8631b2180bdd81b4d9"
     },
     {
       "path": "subset-chart-data.json",
@@ -734,7 +734,7 @@
       "apus-ailab/APUS-OpenJev-v1-4B",
       "apus-ailab/APUS-OpenJev-v1-9B"
     ],
-    "previous_revision": "ac18caa890fde79736bd29734434bb0677b18b10",
+    "previous_revision": "2e393e5517ffc14fc0b400ec11ba319f42ef28bf",
     "weight_files_changed": false
   }
 }

series-downloads.json

@@ -1,7 +1,7 @@
 {
   "schema": "apus-openjev-series-stats.v1",
-  "generated_at": "2026-09-22T03:08:12.998307Z",
-  "observed_at": "2026-09-22T03:08:12.998307Z",
+  "generated_at": "2026-09-22T03:10:37.132203Z",
+  "observed_at": "2026-09-22T03:10:37.132203Z",
   "published_repo_ids": [
     "apus-ailab/APUS-OpenJev-v1-4B",
     "apus-ailab/APUS-OpenJev-v1-9B"