KDL-Frontier-Parser-nano
A 1.2B-parameter open-weight document parsing model, packaged and orchestrated by KoreaDeep as the nano tier of the KDL Frontier Parser family.
ParseBench results
Measured 2026-06-10 with the official ParseBench harness, full set, single end-to-end pass (2,553 test cases, 0 inference failures):
| Dimension | Metric | Score |
|---|---|---|
| Overall (mean) | mean of 5 dimensions | 76.48 |
| Tables | grits_trm_composite | 84.56 |
| Visual Grounding / Layout | rule_pass_rate | 81.83 |
| Content Faithfulness | content_faithfulness | 86.63 |
| Semantic Formatting | normalized_text_score | 66.32 |
| Charts | chart_data_point | 63.08 |
Serving
vllm serve <this-repo> \
--served-model-name kdl-frontier-parser-nano \
--max-model-len 8192 \
--gpu-memory-utilization 0.85 \
--max-num-seqs 24 \
--trust-remote-code \
--limit-mm-per-prompt '{"image":1}'
Benchmark methodology
The ParseBench score is an end-to-end pipeline measurement — this model served via vLLM plus deterministic rule-based post-processing of model output — consistent with how all ParseBench providers are evaluated (every provider is a submitter-hosted endpoint). No other learned models, classifiers, or ensembles are involved: single model, single pass.
About
Built by KoreaDeep, a document-AI company. The larger KDL-Frontier-Parser-ultra is available through DEEP Agent.
- Downloads last month
- 339
Safetensors
Model size
1B params
Tensor type
BF16
·
Evaluation results
- llamaindex/ParseBench leaderboard
- Mean View evaluation results source Pipeline name: kdl_frontier_nano (standalone provider, run-llama/ParseBench PR #49). Full-set single-pass run (2,553 test cases, 0 failures): vLLM serving of the public weights + the deterministic provider pipeline. Layout dimension = layout_element_rule_pass_rate (official leaderboard metric), as re-measured by the ParseBench maintainers on this provider's predictions with the Chart/Flowchart label mapping (run-llama/ParseBench PR #49). Reproducible end-to-end with the provider in the PR.76.36 *
- Table View evaluation results source Pipeline name: kdl_frontier_nano (standalone provider, run-llama/ParseBench PR #49)85.56 *
- Layout View evaluation results source Pipeline name: kdl_frontier_nano (standalone provider, run-llama/ParseBench PR #49)78.84 *
