Category

Long Context: RL environments and AI benchmarks

AI benchmarks
40
Shares entries with
0 categories

Snapshot 2026-10-11. How categories are built · Where the data comes from

In short

What are RL environments for long context?
A simulated workplace where an AI agent practises a task (a code repository, a browser, business software) and an automatic grader scores each attempt. Labs train on thousands of these attempts. Bought to improve models.

Only benchmarks use the Long Context tag: the RL environment directories file their vendors under other categories. Browse all RL environments.

AI benchmarks 40

None confirmed yet.

40 not yet confirmedListed in the catalog; no line from its own site or paper confirms it yet.
  1. CL-bench
  2. LOCA-Bench
  3. Graphwalks BFS 1M F1
  4. Graphwalks BFS 256k F1
  5. Graphwalks Parents 1M F1
  6. Graphwalks Parents 256k F1
  7. OpenAI MRCR v2 8-needle 128K-256K
  8. OpenAI MRCR v2 8-needle 16K-32K
  9. OpenAI MRCR v2 8-needle 256K-512K
  10. OpenAI MRCR v2 8-needle 32K-64K
  11. OpenAI MRCR v2 8-needle 4K-8K
  12. OpenAI MRCR v2 8-needle 512K-1M
  13. OpenAI MRCR v2 8-needle 64K-128K
  14. OpenAI MRCR v2 8-needle 8K-16K
  15. MRCR-v2 128k
  16. MRCR v2 256K
  17. Beam128K
  18. ATLAS
  19. Fiction.LiveBench
  20. ConStory-Bench
  21. BABILong
  22. LongBench Pro
  23. LongVideoBench
  24. STAGE
  25. Beyond Static Dialogues
  26. On Stable Long-Form Generation
  27. needle-1M-bench
  28. MRCR v2
  29. CorpusQA
  30. ES-MemEval
  31. EvolMem
  32. FlenQA
  33. OpenAI-MRCR: 2 needle 128k
  34. (long-context agent monitoring benchmark — not explicitly named in abstract)
  35. Evaluating Long-Horizon Memory for Multi-Party Collaborative Dialogues
  36. extended needle-in-a-haystack benchmark
  37. OpenAI-MRCR: 2 needle 1M
  38. PaperAudit-Bench
  39. Qasper
  40. SQuALITY

Next: pick another domain. Tags come from the RL Research daily scrape, canonicalized nightly.