feat(olmocr): add chandra-ocr-2 runner#3
Open
Khurdhula-Harshavardhan wants to merge 1 commit into
Open
Conversation
Mirrors the existing per-provider olmOCR runners (openai_mini, grok, etc.) but dispatches each rendered PDF page to a self-hosted `datalab-to/chandra-ocr-2` vLLM endpoint over HTTPS. Auth via the `x-api-admin-key` header; runner reads `CHANDRA_MODAL_URL` and `CHANDRA_MODAL_ADMIN_KEY` from env. Request payload uses `task="ocr_layout"` and `temperature=0.0` to stay apples-to-apples with the other candidates (no reasoning/thinking mode — Chandra OCR 2 is a fine-tuned OCR VLM with none). Harness supports `--sample`, `--skip-generation`, `--generate-only`, and `--limit N` (smoke). `RATE_LIMIT=50` matches the throughput of a 10-H100 deployment; turn down if running against a single warm container. README updated under the olmOCR section to list the new runner and the two required env vars.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
olmocr_bench_chandra_ocr2.py+bench/runners/run_chandra_ocr2.py, following the sameopenai_mini/grok/reductotemplatedatalab-to/chandra-ocr-2vLLM endpoint (HTTPS POST/ocr), auth viax-api-admin-keyheadertemperature=0,task="ocr_layout", no reasoning/thinking mode (Chandra OCR 2 is a fine-tuned OCR VLM with none)--sample,--skip-generation,--generate-only,--limit N(smoke)Required env
Notes
RATE_LIMIT=50matches a 10-H100 deployment; turn down if running against a single warm containerTest plan
python -m py_compile