docs: add batch processing speed claim (100+ pages/sec on 8+ cores)

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

Author: bundolee

README.md

@@ -494,7 +494,7 @@ Yes. For digital PDFs, text extraction works out of the box. For scanned PDFs, u
 
 ### How fast is it?
 
-Local mode processes 20+ pages per second on CPU (0.05s/page). Hybrid mode processes 2+ pages per second (0.43s/page) with significantly higher accuracy for complex documents. No GPU required. Benchmarked on Apple M4. [Full benchmark details](https://github.com/opendataloader-project/opendataloader-bench)
+Local mode processes 20+ pages per second on CPU (0.05s/page). Hybrid mode processes 2+ pages per second (0.43s/page) with significantly higher accuracy for complex documents. No GPU required. Benchmarked on Apple M4. [Full benchmark details](https://github.com/opendataloader-project/opendataloader-bench). With multi-process batch processing, throughput exceeds 100 pages per second on 8+ core machines.
 
 ### Does it handle multi-column layouts?
 

content/docs/faq.mdx

@@ -191,10 +191,7 @@ Yes. OpenDataLoader:
 
 ### How fast is OpenDataLoader?
 
-OpenDataLoader processes **20+ pages per second** on typical hardware (CPU only, no GPU). See our [benchmark results](/docs/benchmark) for details. Exact speed depends on:
-- Document complexity
-- Number of tables/images
-- Output formats selected
+Local mode processes 20+ pages per second on CPU (0.05s/page). Hybrid mode processes 2+ pages per second (0.43s/page) with significantly higher accuracy for complex documents. No GPU required. Benchmarked on Apple M4. [Full benchmark details](https://github.com/opendataloader-project/opendataloader-bench). With multi-process batch processing, throughput exceeds 100 pages per second on 8+ core machines.
 
 ### Can I process multiple PDFs at once?
 

content/docs/index.mdx

@@ -7,7 +7,7 @@ OpenDataLoader PDF converts PDFs into **LLM-ready Markdown and JSON** with accur
 
 **Why developers choose OpenDataLoader:**
 - **Deterministic** — Same input always produces same output (no LLM hallucinations)
-- **Fast** — Process 20+ pages per second on CPU
+- **Fast** — Process 20+ pages per second on CPU (100+ with batch parallelism)
 - **Private** — 100% local, zero data transmission
 - **Accurate** — Bounding boxes for every element, correct multi-column reading order