RotomArenaAI gaming benchmark • 2 benchmarks

Matches · 2026-10-11 · rules version c4

GPT-OSS 20B vs Bonsai 2 27B

Void. gpt_oss_20b was on autopilot for 100% of its turns. It does not count toward the ratings.
Player 1 · winner
GPT-OSS 20B
GPT-OSS 20B's team
autopilot 100%
Player 2
Bonsai 2 27B
Bonsai 2 27B's team
autopilot 100%

GPT-OSS 20B won in 38 turns · 10 min wall clock

The replay player loads from play.pokemonshowdown.com. The full battle log is below.

Decisions

  1. Team preview
    GPT-OSS 20B brought Rotomwash,toxapex,garchompautopilot
    Incident report
    [API ERROR] LLM request failed after 3 attempt(s): Connection error.. Bringing rotomwash, toxapex, garchomp.
    Bonsai 2 27B brought Garchomp,toxapex,krookodileautopilot
    Incident report
    [API ERROR] LLM request failed after 3 attempt(s): Connection error.. Bringing garchomp, toxapex, krookodile.
  2. Turn 1
    Bonsai 2 27B used Swordsdanceautopilot
    Incident report
    [API ERROR] LLM request failed after 3 attempt(s): Connection error.. Default move played.
    GPT-OSS 20B used Hydropumpautopilot
    Incident report
    [API ERROR] LLM request failed after 3 attempt(s): Connection error.. Default move played.
  3. Turn 2
    GPT-OSS 20B used Hydropumpautopilot
    Incident report
    [API ERROR] LLM request failed after 3 attempt(s): Connection error.. Default move played.
    Bonsai 2 27B used Swordsdanceautopilot
    Incident report
    [API ERROR] LLM request failed after 3 attempt(s): Connection error.. Default move played.
  4. Turn 3
    GPT-OSS 20B used Hydropumpautopilot
    Incident report
    [API ERROR] LLM request failed after 3 attempt(s): Connection error.. Default move played.
    Bonsai 2 27B used Swordsdanceautopilot
    Incident report
    [API ERROR] LLM request failed after 3 attempt(s): Connection error.. Default move played.
  5. Turn 4
    Bonsai 2 27B used Swordsdanceautopilot
    Incident report
    [API ERROR] LLM request failed after 3 attempt(s): Connection error.. Default move played.
    GPT-OSS 20B used Hydropumpautopilot
    Incident report
    [API ERROR] LLM request failed after 3 attempt(s): Connection error.. Default move played.
    Bonsai 2 27B switched to Toxapexautopilot
    Incident report
    [API ERROR] LLM request failed after 3 attempt(s): Connection error.. Default move played.
  6. Turn 5
    GPT-OSS 20B used Hydropumpautopilot
    Incident report
    [API ERROR] LLM request failed after 3 attempt(s): Connection error.. Default move played.
    Bonsai 2 27B used Sludgebombautopilot
    Incident report
    [API ERROR] LLM request failed after 3 attempt(s): Connection error.. Default move played.
  7. Turn 6
    GPT-OSS 20B used Hydropumpautopilot
    Incident report
    [API ERROR] LLM request failed after 3 attempt(s): Connection error.. Default move played.
    Bonsai 2 27B used Sludgebombautopilot
    Incident report
    [API ERROR] LLM request failed after 3 attempt(s): Connection error.. Default move played.
  8. Turn 7
    GPT-OSS 20B used Hydropumpautopilot
    Incident report
    [API ERROR] LLM request failed after 3 attempt(s): Connection error.. Default move played.
    Bonsai 2 27B used Sludgebombautopilot
    Incident report
    [API ERROR] LLM request failed after 3 attempt(s): Connection error.. Default move played.
  9. Turn 8
    GPT-OSS 20B used Hydropumpautopilot
    Incident report
    [API ERROR] LLM request failed after 3 attempt(s): Connection error.. Default move played.
    Bonsai 2 27B used Sludgebombautopilot
    Incident report
    [API ERROR] LLM request failed after 3 attempt(s): Connection error.. Default move played.
  10. Turn 9
    GPT-OSS 20B used Thunderboltautopilot
    Incident report
    [API ERROR] LLM request failed after 3 attempt(s): Connection error.. Default move played.
    Bonsai 2 27B used Sludgebombautopilot
    Incident report
    [API ERROR] LLM request failed after 3 attempt(s): Connection error.. Default move played.
    Bonsai 2 27B switched to Krookodileautopilot
    Incident report
    [API ERROR] LLM request failed after 3 attempt(s): Connection error.. Default move played.
  11. Turn 10
    GPT-OSS 20B used Thunderboltautopilot
    Incident report
    [API ERROR] LLM request failed after 3 attempt(s): Connection error.. Default move played.
    Bonsai 2 27B used Earthquakeautopilot
    Incident report
    [API ERROR] LLM request failed after 3 attempt(s): Connection error.. Default move played.
  12. Turn 11
    GPT-OSS 20B used Thunderboltautopilot
    Incident report
    [API ERROR] LLM request failed after 3 attempt(s): Connection error.. Default move played.
    Bonsai 2 27B used Earthquakeautopilot
    Incident report
    [API ERROR] LLM request failed after 3 attempt(s): Connection error.. Default move played.
  13. Turn 12
    GPT-OSS 20B used Thunderboltautopilot
    Incident report
    [API ERROR] LLM request failed after 3 attempt(s): Connection error.. Default move played.
    Bonsai 2 27B used Earthquakeautopilot
    Incident report
    [API ERROR] LLM request failed after 3 attempt(s): Connection error.. Default move played.
  14. Turn 13
    GPT-OSS 20B used Thunderboltautopilot
    Incident report
    [API ERROR] LLM request failed after 3 attempt(s): Connection error.. Default move played.
    Bonsai 2 27B used Earthquakeautopilot
    Incident report
    [API ERROR] LLM request failed after 3 attempt(s): Connection error.. Default move played.
  15. Turn 14
    GPT-OSS 20B used Thunderboltautopilot
    Incident report
    [API ERROR] LLM request failed after 3 attempt(s): Connection error.. Default move played.
    Bonsai 2 27B used Earthquakeautopilot
    Incident report
    [API ERROR] LLM request failed after 3 attempt(s): Connection error.. Default move played.
  16. Turn 15
    GPT-OSS 20B used Thunderboltautopilot
    Incident report
    [API ERROR] LLM request failed after 3 attempt(s): Connection error.. Default move played.
    Bonsai 2 27B used Earthquakeautopilot
    Incident report
    [API ERROR] LLM request failed after 3 attempt(s): Connection error.. Default move played.
  17. Turn 16
    GPT-OSS 20B used Thunderboltautopilot
    Incident report
    [API ERROR] LLM request failed after 3 attempt(s): Connection error.. Default move played.
    Bonsai 2 27B used Earthquakeautopilot
    Incident report
    [API ERROR] LLM request failed after 3 attempt(s): Connection error.. Default move played.
  18. Turn 17
    GPT-OSS 20B used Thunderboltautopilot
    Incident report
    [API ERROR] LLM request failed after 3 attempt(s): Connection error.. Default move played.
    Bonsai 2 27B used Earthquakeautopilot
    Incident report
    [API ERROR] LLM request failed after 3 attempt(s): Connection error.. Default move played.
  19. Turn 18
    GPT-OSS 20B used Thunderboltautopilot
    Incident report
    [API ERROR] LLM request failed after 3 attempt(s): Connection error.. Default move played.
    Bonsai 2 27B used Earthquakeautopilot
    Incident report
    [API ERROR] LLM request failed after 3 attempt(s): Connection error.. Default move played.
  20. Turn 19
    Bonsai 2 27B used Earthquakeautopilot
    Incident report
    [API ERROR] LLM request failed after 3 attempt(s): Connection error.. Default move played.
    GPT-OSS 20B used Thunderboltautopilot
    Incident report
    [API ERROR] LLM request failed after 3 attempt(s): Connection error.. Default move played.
  21. Turn 20
    GPT-OSS 20B used Thunderboltautopilot
    Incident report
    [API ERROR] LLM request failed after 3 attempt(s): Connection error.. Default move played.
    Bonsai 2 27B used Earthquakeautopilot
    Incident report
    [API ERROR] LLM request failed after 3 attempt(s): Connection error.. Default move played.
  22. Turn 21
    GPT-OSS 20B used Thunderboltautopilot
    Incident report
    [API ERROR] LLM request failed after 3 attempt(s): Connection error.. Default move played.
    Bonsai 2 27B used Earthquakeautopilot
    Incident report
    [API ERROR] LLM request failed after 3 attempt(s): Connection error.. Default move played.
  23. Turn 22
    GPT-OSS 20B used Thunderboltautopilot
    Incident report
    [API ERROR] LLM request failed after 3 attempt(s): Connection error.. Default move played.
    Bonsai 2 27B used Stoneedgeautopilot
    Incident report
    [API ERROR] LLM request failed after 3 attempt(s): Connection error.. Default move played.
  24. Turn 23
    GPT-OSS 20B used Thunderboltautopilot
    Incident report
    [API ERROR] LLM request failed after 3 attempt(s): Connection error.. Default move played.
    Bonsai 2 27B used Stoneedgeautopilot
    Incident report
    [API ERROR] LLM request failed after 3 attempt(s): Connection error.. Default move played.
  25. Turn 24
    GPT-OSS 20B used Thunderboltautopilot
    Incident report
    [API ERROR] LLM request failed after 3 attempt(s): Connection error.. Default move played.
    Bonsai 2 27B used Stoneedgeautopilot
    Incident report
    [API ERROR] LLM request failed after 3 attempt(s): Connection error.. Default move played.
  26. Turn 25
    GPT-OSS 20B used Voltswitchautopilot
    Incident report
    [API ERROR] LLM request failed after 3 attempt(s): Connection error.. Default move played.
    Bonsai 2 27B used Stoneedgeautopilot
    Incident report
    [API ERROR] LLM request failed after 3 attempt(s): Connection error.. Default move played.
    GPT-OSS 20B switched to Toxapexautopilot
    Incident report
    [API ERROR] LLM request failed after 3 attempt(s): Connection error.. Default move played.
  27. Turn 26
    Bonsai 2 27B used Stoneedgeautopilot
    Incident report
    [API ERROR] LLM request failed after 3 attempt(s): Connection error.. Default move played.
    GPT-OSS 20B used Banefulbunkerautopilot
    Incident report
    [API ERROR] LLM request failed after 3 attempt(s): Connection error.. Default move played.
  28. Turn 27
    GPT-OSS 20B used Banefulbunkerautopilot
    Incident report
    [API ERROR] LLM request failed after 3 attempt(s): Connection error.. Default move played.
    Bonsai 2 27B used Stoneedgeautopilot
    Incident report
    [API ERROR] LLM request failed after 3 attempt(s): Connection error.. Default move played.
  29. Turn 28
    GPT-OSS 20B used Banefulbunkerautopilot
    Incident report
    [API ERROR] LLM request failed after 3 attempt(s): Connection error.. Default move played.
    Bonsai 2 27B used Stoneedgeautopilot
    Incident report
    [API ERROR] LLM request failed after 3 attempt(s): Connection error.. Default move played.
  30. Turn 29
    GPT-OSS 20B used Banefulbunkerautopilot
    Incident report
    [API ERROR] LLM request failed after 3 attempt(s): Connection error.. Default move played.
    Bonsai 2 27B used Stoneedgeautopilot
    Incident report
    [API ERROR] LLM request failed after 3 attempt(s): Connection error.. Default move played.
  31. Turn 30
    GPT-OSS 20B used Banefulbunkerautopilot
    Incident report
    [API ERROR] LLM request failed after 3 attempt(s): Connection error.. Default move played.
    Bonsai 2 27B used Knockoffautopilot
    Incident report
    [API ERROR] LLM request failed after 3 attempt(s): Connection error.. Default move played.
  32. Turn 31
    GPT-OSS 20B used Banefulbunkerautopilot
    Incident report
    [API ERROR] LLM request failed after 3 attempt(s): Connection error.. Default move played.
    Bonsai 2 27B used Knockoffautopilot
    Incident report
    [API ERROR] LLM request failed after 3 attempt(s): Connection error.. Default move played.
  33. Turn 32
    GPT-OSS 20B used Banefulbunkerautopilot
    Incident report
    [API ERROR] LLM request failed after 3 attempt(s): Connection error.. Default move played.
    Bonsai 2 27B used Knockoffautopilot
    Incident report
    [API ERROR] LLM request failed after 3 attempt(s): Connection error.. Default move played.
  34. Turn 33
    GPT-OSS 20B used Banefulbunkerautopilot
    Incident report
    [API ERROR] LLM request failed after 3 attempt(s): Connection error.. Default move played.
    Bonsai 2 27B used Knockoffautopilot
    Incident report
    [API ERROR] LLM request failed after 3 attempt(s): Connection error.. Default move played.
  35. Turn 34
    GPT-OSS 20B used Recoverautopilot
    Incident report
    [API ERROR] LLM request failed after 3 attempt(s): Connection error.. Default move played.
    Bonsai 2 27B used Knockoffautopilot
    Incident report
    [API ERROR] LLM request failed after 3 attempt(s): Connection error.. Default move played.
  36. Turn 35
    GPT-OSS 20B used Recoverautopilot
    Incident report
    [API ERROR] LLM request failed after 3 attempt(s): Connection error.. Default move played.
    Bonsai 2 27B used Knockoffautopilot
    Incident report
    [API ERROR] LLM request failed after 3 attempt(s): Connection error.. Default move played.
  37. Turn 36
    GPT-OSS 20B used Recoverautopilot
    Incident report
    [API ERROR] LLM request failed after 3 attempt(s): Connection error.. Default move played.
    Bonsai 2 27B used Knockoffautopilot
    Incident report
    [API ERROR] LLM request failed after 3 attempt(s): Connection error.. Default move played.
  38. Turn 37
    GPT-OSS 20B used Recoverautopilot
    Incident report
    [API ERROR] LLM request failed after 3 attempt(s): Connection error.. Default move played.
    Bonsai 2 27B used Knockoffautopilot
    Incident report
    [API ERROR] LLM request failed after 3 attempt(s): Connection error.. Default move played.
  39. Turn 38
    GPT-OSS 20B used Recoverautopilot
    Incident report
    [API ERROR] LLM request failed after 3 attempt(s): Connection error.. Default move played.
    Bonsai 2 27B used Knockoffautopilot
    Incident report
    [API ERROR] LLM request failed after 3 attempt(s): Connection error.. Default move played.