This release adds 3 notable features for engineering teams evaluating rollout.
✓ No known CVEs patched in this version
Topics
+7 more
Summary
AI summaryUpdates 2026.6.11 v3.7.0 released, 2026.6.11 v3.7.0 发布, and OpenVINO across a mixed release.
Changes in this release
| Type | Severity | Summary | CVE |
|---|---|---|---|
| Feature | Medium |
Adds 50‑language unified support in a single model. Adds 50‑language unified support in a single model. Source: llm_adapter@2026-06-11 Confidence: high |
— |
| Feature | Low |
Introduces three model tiers: tiny (1.5 M), small (7.7 M), medium (34.5 M) for edge, mobile, and server deployment. Introduces three model tiers: tiny (1.5 M), small (7.7 M), medium (34.5 M) for edge, mobile, and server deployment. Source: llm_adapter@2026-06-11 Confidence: high |
— |
| Feature | Low |
Enhances specialized scenarios: digital displays, dot‑matrix characters, tire prints, and industrial text recognition. Enhances specialized scenarios: digital displays, dot‑matrix characters, tire prints, and industrial text recognition. Source: llm_adapter@2026-06-11 Confidence: high |
— |
| Performance | Medium |
Improves medium‑tier detection accuracy by +4.6% and recognition accuracy by +5.1%. Improves medium‑tier detection accuracy by +4.6% and recognition accuracy by +5.1%. Source: llm_adapter@2026-06-11 Confidence: high |
— |
| Performance | Medium |
Accelerates medium‑tier CPU inference 5.2× with OpenVINO. Accelerates medium‑tier CPU inference 5.2× with OpenVINO. Source: llm_adapter@2026-06-11 Confidence: high |
— |
| Performance | Medium |
Accelerates tiny‑tier inference 6.1× on Apple M4 (tiny). Accelerates tiny‑tier inference 6.1× on Apple M4 (tiny). Source: llm_adapter@2026-06-11 Confidence: high |
— |
| Performance | Medium |
Reduces medium‑tier inference latency to 0.13 s on A100 GPU. Reduces medium‑tier inference latency to 0.13 s on A100 GPU. Source: llm_adapter@2026-06-11 Confidence: high |
— |
Full changelog
2026.6.11 v3.7.0 released
-
Release PP-OCRv6
- Accuracy boost: Medium tier achieves +4.6% detection and +5.1% recognition over PP-OCRv5_server, surpassing mainstream VLMs (Qwen3-VL-235B, GPT-5.5) with only 34.5M parameters.
- 50 languages unified: Single model covers Chinese, English, Japanese, and 46 Latin-script languages — no model switching needed.
- Specialized scenarios: Major improvements in digital displays, dot-matrix characters, tire prints, and industrial text recognition.
- Faster inference: 5.2× CPU speedup (OpenVINO), 6.1× on Apple M4 (tiny), 0.13s on A100 GPU.
- Three tiers for all scenarios: tiny (1.5M) / small (7.7M) / medium (34.5M) for edge, mobile, and server deployment.
2026.6.11 v3.7.0 发布
-
发布 PP-OCRv6
- 精度全面提升：medium 档相比 PP-OCRv5_server 检测精度提升 4.6%、识别精度提升 5.1%，以仅 34.5M 参数超越 Qwen3-VL-235B、GPT-5.5 等主流视觉语言大模型。
- 50 种语言统一支持：单一模型覆盖中文、英文、日文及 46 种拉丁语系语言，无需为不同语种切换模型。
- 专业场景增强：数码显示屏、点阵字符、轮胎印字、工业字符等传统 VLM 难以覆盖的场景识别能力大幅提升。
- 推理速度更快：medium 档 CPU OpenVINO 推理加速 5.2×，tiny 档 Apple M4 加速 6.1×，A100 上仅需 0.13s。
- 三档模型覆盖全场景：tiny（1.5M）/ small（7.7M）/ medium（34.5M）分别面向端侧/移动端/服务端部署。
Full Changelog: https://github.com/PaddlePaddle/PaddleOCR/compare/v3.6.0...v3.7.0
Weekly OSS security release digest.
The CVE patches and breaking changes that affected production tools this week. One email, every Sunday.
No spam, unsubscribe anytime.
Share this release
About PaddleOCR
All releases →Beta — feedback welcome: [email protected]