{"type":"analysis","id":"analysis:research-open-weight-extractors-hold-on-short-pages-then-lose-the-long-lists","slug":"research-open-weight-extractors-hold-on-short-pages-then-lose-the-long-lists","title":"Open-weight extractors hold on short pages, then lose the long lists","description":"LlamaIndex ExtractBench does evaluate self-hosted open models on schema-guided document extraction. Published scores put Qwen-class models near 90 percent overall value F1, then show a sharp completeness drop on documents longer than fifty pages, while LlamaExtract still leads a first-party leaderboard.","published_at":"2026-10-25T12:00:00.000Z","updated_at":"2026-10-25T12:00:00.000Z","record_date":"2026-10-25","topics":["agents","coding_agents","github","research"],"source_urls":["https://arxiv.org/abs/2607.29677","https://huggingface.co/datasets/llamaindex/ExtractBench","https://www.extractbench.ai/","https://www.llamaindex.ai/blog/introducing-extractbench","https://raw.githubusercontent.com/run-llama/ExtractBench/main/leaderboard.csv","https://raw.githubusercontent.com/run-llama/ExtractBench/main/README.md","https://www.datalab.to/blog/extractbench-scoring-bug","https://developers.llamaindex.ai/llamaparse/general/pricing/","https://arxiv.org/abs/2602.12247","https://huggingface.co/Qwen/Qwen3.6-35B-A3B-FP8"],"language":"en","routes":{"html":"https://newruntime.com/analysis/research-open-weight-extractors-hold-on-short-pages-then-lose-the-long-lists/","markdown":"https://newruntime.com/analysis/research-open-weight-extractors-hold-on-short-pages-then-lose-the-long-lists.md","json":"https://newruntime.com/analysis/research-open-weight-extractors-hold-on-short-pages-then-lose-the-long-lists.json"},"author":{"role":"author","name":"Andrey Reshetnikov","profile":"https://newruntime.com/owner-profile.md","profile_json":"https://newruntime.com/owner-profile.json","same_as":["https://www.linkedin.com/in/reshetnikov1/","https://t.me/lapetuse"]},"editorial_provenance":{"desk":"authorial","target_ref":"deep_research:d12776d4-4565-467b-8fbd-2f7786eaac43","authorial_article_id":"d12776d4-4565-467b-8fbd-2f7786eaac43"},"next_reads":[{"type":"topic","path":"/topics/agents/","url":"https://newruntime.com/topics/agents/","title":"Agents","reason":"Explore the agents topic hub.","media_type":"text/html","markdown_url":"https://newruntime.com/topics/agents.md","json_url":"https://newruntime.com/topics/agents.json"},{"type":"topic","path":"/topics/coding-agents/","url":"https://newruntime.com/topics/coding-agents/","title":"Coding Agents","reason":"Explore the coding_agents topic hub.","media_type":"text/html","markdown_url":"https://newruntime.com/topics/coding-agents.md","json_url":"https://newruntime.com/topics/coding-agents.json"},{"type":"topic","path":"/topics/github/","url":"https://newruntime.com/topics/github/","title":"Github","reason":"Explore the github topic hub.","media_type":"text/html","markdown_url":"https://newruntime.com/topics/github.md","json_url":"https://newruntime.com/topics/github.json"},{"type":"collection","path":"/analysis/","url":"https://newruntime.com/analysis/","title":"Analysis","reason":"Open the latest authorial analysis.","media_type":"text/html","json_url":"https://newruntime.com/analysis.json"},{"type":"collection","path":"/source-ledger.json","url":"https://newruntime.com/source-ledger.json","title":"Source Ledger","reason":"Inspect the public source evidence graph.","media_type":"application/json"}]}