All AI models · LLM benchmarks · Methodology

LightOnOCR 2

LightOnOCR 2 hosted by IONOS in Berlin, Germany. Zero data retention.

Model facts

Context window
33K tokens
Maximum output
8K tokens
Cheapest paid input
$0.178 per 1M tokens
Cheapest paid output
$0.346 per 1M tokens
Open weights
yes
Providers
1

Capabilities and modalities

tool calling, attachments, open weights, text, image.

Local hardware estimate

At 4K context: Q4_K_M 1.6 GiB, Q8_0 2.1 GiB, F16 2.9 GiB working memory.

Parameters
1.0 billion
Architecture
lightonocr
Model license
apache-2.0

Planning estimate adapted from the MIT-licensed whichllm estimator using metadata from Hugging Face; not a vendor minimum requirement.

Observed price history

  • 2026-09-11: $0.178 input / $0.346 output per 1M tokens

API providers

  • NanoGPT — model id lightonai/LightOnOCR-2-1B; $0.178 input / $0.346 output per 1M tokens; provider documentation