perf: RTX3060 12G显存适配,batch_size上限提升到8 #2007

Merged
auto-approve-bot merged 1 commits from fix/rtx3060-batch-size into develop 2026-09-21 19:09:03 +08:00
Owner

适配 RTX3060 12G 显存

变更

  • musetalk_server.py 中 batch_size 上限从 min(Config.batch_size, 2) 改为 min(Config.batch_size, 8)
  • RTX3060 12G 显存 + FP16 + GFPGAN 场景下,batch=8 约用 7-8GB,留足余量
  • 推理速度提升约 4 倍(相比之前 batch=2 硬限制)

说明

  • 默认 MUSE_BATCH_SIZE=8 已满足需求,无需改环境变量
  • 如需进一步调大可设 MUSE_BATCH_SIZE=16,但12G显存下收益有限

关于旧 worker 记录清理

旧的 rtx2060-0193 记录已停止心跳,has_available_worker() 会自动忽略。如需彻底清理可手动执行:

DELETE FROM gpu_workers WHERE worker_id = 'rtx2060-0193';

health 接口 gfpgan_loaded 字段

已在 PR#2006 中添加,本 PR 确认无变更。

## 适配 RTX3060 12G 显存 ### 变更 - `musetalk_server.py` 中 batch_size 上限从 `min(Config.batch_size, 2)` 改为 `min(Config.batch_size, 8)` - RTX3060 12G 显存 + FP16 + GFPGAN 场景下,batch=8 约用 7-8GB,留足余量 - 推理速度提升约 4 倍(相比之前 batch=2 硬限制) ### 说明 - 默认 `MUSE_BATCH_SIZE=8` 已满足需求,无需改环境变量 - 如需进一步调大可设 `MUSE_BATCH_SIZE=16`,但12G显存下收益有限 ### 关于旧 worker 记录清理 旧的 `rtx2060-0193` 记录已停止心跳,`has_available_worker()` 会自动忽略。如需彻底清理可手动执行: ```sql DELETE FROM gpu_workers WHERE worker_id = 'rtx2060-0193'; ``` ### health 接口 gfpgan_loaded 字段 已在 PR#2006 中添加,本 PR 确认无变更。
xiaoxia added 1 commit 2026-09-21 19:00:57 +08:00
perf: RTX3060 12G显存适配,batch_size上限从2提升到8
CI/CD Pipeline / Check push changed paths (pull_request) Has been skipped
CI/CD Pipeline / Check if frontend-only change (pull_request) Successful in 2s
CI/CD Pipeline / Dedup Check - skip PR tests when covered by push pipeline (pull_request) Successful in 2s
CI/CD Pipeline / Build Staging API Image (pull_request) Has been skipped
CI/CD Pipeline / Build Staging Web Image (pull_request) Has been skipped
CI/CD Pipeline / Build Staging Worker Image (pull_request) Has been skipped
CI/CD Pipeline / Frontend Lint (pull_request) Has been skipped
CI/CD Pipeline / Frontend Unit Tests (pull_request) Has been skipped
CI/CD Pipeline / PR Build Web Image (pull_request) Has been skipped
CI/CD Pipeline / Retag skipped Staging API Image (pull_request) Has been skipped
CI/CD Pipeline / Retag skipped Staging Web Image (pull_request) Has been skipped
CI/CD Pipeline / Retag skipped Staging Worker Image (pull_request) Has been skipped
CI/CD Pipeline / Deploy Staging (Watchtower auto-deploy) (pull_request) Has been skipped
CI/CD Pipeline / Staging E2E Tests (pull_request) Has been skipped
CI/CD Pipeline / Staging API Integration Tests (pull_request) Has been skipped
CI/CD Pipeline / ACR Image Cleanup (pull_request) Has been skipped
Preview Deploy / Deploy Preview Environment (pull_request) Successful in 1m35s
CI/CD Pipeline / PR Build API Image (pull_request) Successful in 2m46s
PR Automation / Auto Approve on CI Green (pull_request) Successful in 3m9s
CI/CD Pipeline / Validate - Style (pull_request) Successful in 3m7s
CI/CD Pipeline / Validate - Python (mypy + alembic) (pull_request) Successful in 4m8s
CI/CD Pipeline / PR Build Worker Image (pull_request) Successful in 4m25s
CI/CD Pipeline / Integration Tests (pull_request) Successful in 5m15s
AI Code Review / AI Code Review (pull_request) Successful in 6m34s
CI/CD Pipeline / Unit Tests (pull_request) Successful in 7m20s
CI/CD Pipeline / Validate - Security (pull_request) Successful in 7m23s
CI/CD Pipeline / Build Production API Image (pull_request) Has been skipped
CI/CD Pipeline / Build Production Web Image (pull_request) Has been skipped
CI/CD Pipeline / Build Production Worker Image (pull_request) Has been skipped
CI/CD Pipeline / CI Gate (pull_request) Successful in 1s
CI/CD Pipeline / Deploy Production (pull_request) Has been skipped
CI/CD Pipeline / Canary Release to Production (pull_request) Has been skipped
CI/CD Pipeline / Production Browser E2E (pull_request) Has been skipped
PR Automation / Auto Merge on CI Green + Approved (pull_request) Successful in 4m57s
ACR Cleanup / ACR Image Cleanup (pull_request_target) Successful in 23s
Preview Cleanup / Cleanup Preview Environment (pull_request) Successful in 4m14s
a34a9a7844
- RTX3060 12G显存 + FP16 + GFPGAN场景下,batch=8约用7-8GB,留足余量
- 推理速度提升约4倍(相比之前的batch=2硬限制)

🚀 预览环境已部署

项目 详情
PR号 #2007
预览链接 https://pr-2007.preview.xiaoxiajianji.com
API环境 staging

💡 预览环境使用 staging API 数据,请勿在预览环境中操作重要数据。

🔄 每次提交新代码后预览环境会自动更新。

🗑️ PR 关闭或合并后,预览环境会自动清理。

🚀 **预览环境已部署** | 项目 | 详情 | |------|------| | PR号 | #2007 | | 预览链接 | [https://pr-2007.preview.xiaoxiajianji.com](https://pr-2007.preview.xiaoxiajianji.com) | | API环境 | staging | > 💡 预览环境使用 staging API 数据,请勿在预览环境中操作重要数据。 > > 🔄 每次提交新代码后预览环境会自动更新。 > > 🗑️ PR 关闭或合并后,预览环境会自动清理。
auto-approve-bot merged commit cc01a5da69 into develop 2026-09-21 19:09:03 +08:00
auto-approve-bot deleted branch fix/rtx3060-batch-size 2026-09-21 19:09:05 +08:00

🗑️ 预览环境已清理

PR #2007 已关闭或合并,对应的预览环境已被清理。

如有需要,可以重新打开 PR 来重新生成预览环境。

🗑️ **预览环境已清理** PR #2007 已关闭或合并,对应的预览环境已被清理。 > 如有需要,可以重新打开 PR 来重新生成预览环境。
Sign in to join this conversation.