TrOCR Base Printed is an open-source vision model created by Microsoft for optical character recognition. It is an encoder-decoder Transformer model that takes in images and outputs recognized text.
Input
Output
Context
-
Max Output
50K
Parameters
333.3M
Input Modalities
Output Modalities
Some values come from web research and may be incorrect. Report incorrect data
None the catalog can show from stated facts or measured scores.
Download 1.33 GB · 1 GGUF builds
| Precision | Weights | One request at 4K tokens | Smallest GPU that fits |
|---|---|---|---|
| FP16 / BF16 | 0.62 GB | 0.98 GB | 1x NVIDIA T4 (16 GB) |
| INT8 | 0.31 GB | 0.65 GB | 1x NVIDIA T4 (16 GB) |
| INT4 | 0.16 GB | 0.49 GB | 1x NVIDIA T4 (16 GB) |
Computed by Inferbase from the 333,331,200 parameters in the creator's Hugging Face repository; one request adds its KV cache, activations and framework overhead, with the layer layout estimated from the parameter count.
Public resources
Specification sources
As of Nov 11, 2026.
Answered from the facts on this page, with their sources.
Microsoft has not published a context window for TrOCR Base Printed. The catalog does not estimate one.
Yes. TrOCR Base Printed is an open-weight model under the MIT license. You can download and self-host it within the license terms.
TrOCR Base Printed has 333,331,200 parameters (333.3M), as stated in the creator's Hugging Face repository.
TrOCR Base Printed accepts Image and produces Text. The Capabilities card lists the API features it supports, such as function calling and structured output.
No benchmark source has published a result for TrOCR Base Printed yet, so it is shown as unmeasured, not estimated.
Not on Inferbase Managed today. Add your own provider key or endpoint that serves TrOCR Base Printed, and Inferbase runs it on your account with the same routing and logs.
The weights take 0.6 GB at FP16 / BF16, 0.3 GB at INT8, 0.2 GB at INT4, computed from the 333,331,200 parameters its creator publishes. Each request adds its KV cache and the framework's overhead. The hardware section shows one request and the smallest GPU that fits; the GPU planner sizes it for your context length and traffic.
Start on the free plan with 5,000 requests a month, no card required.