VQV Signal
According to a post on X by @OpenBMB, the 1.3B MiniCPM-OCR model is now supported in vLLM v0.33.0. Docling v2.149.0 added a minicpm_ocr preset for English and Chinese OCR, which the post says runs on a single consumer GPU. OpenBMB thanked both projects. The new element is that the model is available through two widely used document and serving tools.
According to a post on X by @OpenBMB, the 1.3B MiniCPM-OCR model is now supported in vLLM v0.33.0. Docling v2.149.0 added a minicpm_ocr preset for English and Chinese OCR, which the post says runs on a single consumer GPU. OpenBMB thanked both projects. The new element is that the model is available through two widely used document and serving tools.
AI-assisted summary based on the listed source.
vLLM v0.33.0 now supports the 1.3B MiniCPM-OCR model, and Docling v2.149.0 added a minicpm_ocr preset for English and Chinese OCR on a single consumer GPU. OpenBMB thanked both projects. Why it matters: A small, open OCR model is now one flag away in two widely used document and serving tools. Also:...
Teams running document processing or model serving may be able to test this OCR model within existing vLLM and Docling setups. If it runs on a single consumer GPU as described, it could be a lower-cost option for English and Chinese documents. The post does not include accuracy or speed benchmarks, so independent testing will be the next thing to watch.
Hardware and robotics watchers may want to track whether this becomes a product, benchmark, or deployment signal.
VQV organizes public signals from inspectable sources. It does not independently verify the underlying report.
Signal Strength reflects source quality, relevance, freshness and evidence. Public Interest helps organize discovery; it is not proof of truth.
VQV surfaced this signal because it is recent, relevant to LLM Inference, connected to Scobleizer News Pulse (X).
No login, cookies, social SDKs, or automatic posting.
