Spaces:

qpluslab
/

OpenRA-Bench

Running

App Files Files Community

OpenRA-Bench / openra_bench

4.13 MB

Ctrl+K

Ctrl+K

9 contributors

History: 447 commits

Xiaochuang Yuan

combat-prevent-retreat: per-tier maps reinforce encirclement geometry

dac8bc3 3 months ago

_vendor
Vendor training prompt/briefing/minimap v2 (byte-identical) 3 months ago
scenarios
combat-prevent-retreat: per-tier maps reinforce encirclement geometry 3 months ago
__init__.py

290 Bytes
Add provider-agnostic model agent (vLLM/OpenRouter/Bedrock) 3 months ago
adversarial.py

2.66 kB
Adversarial 1v1 spotlight: ladder family + rating + Elo wiring 3 months ago
agent.py

36.7 kB
Phase 1 engine audit: ENGINE_AUDIT.md + bench-side closures 3 months ago
battle_viewer.py

5.84 kB
Training-parity minimap (real terrain + legend) + viewer (system/thinking/debrief) 3 months ago
botgen.py

2.43 kB
feat(engine+bench): raider bot — worker-priority harasser (Rust ScriptedBehavior + bench registration) 3 months ago
controller.py

6.16 kB
Phase 1: unified Controller interface for the eval stack 3 months ago
eval_core.py

24.9 kB
Defensive fix: strip agent.cash=0 when starting_cash>0 in tmp YAML 3 months ago
full_playback.py

14.1 kB
Phase 4 data plumbing + 3rd defect partial fix 3 months ago
game_knowledge.py

22.4 kB
Bucket B: glossary entries for mslo / heli / tanya / dd 3 months ago
goal_tracker.py

5.26 kB
Strategy packs: faithful 'destroy key economic buildings' objective 3 months ago
handoff.py

6.15 kB
Add handoff ablation (recover-from-deficit / capitalize-on-advantage) 3 months ago
human_labeling.py

38 kB
Land PR #15 surgical fixes (Windows safety + human-play hints) 3 months ago
human_study.py

4.61 kB
Paper-experiment prep: human-study harness + pass^k + paper plan 3 months ago
leaderboard.py

5.33 kB
win-speed bonus: record + reward how fast a model wins 3 months ago
mapgen.py

13.4 kB
mapgen: obstacles, bridges-arena, chokepoint-arena + per-scenario authoring 3 months ago
minimap.py

22.1 kB
Merge origin/main into pr13-revised 3 months ago
one_v_one.py

6.79 kB
Phase 3: 1v1 full-macro adversarial harness 3 months ago
pairwise.py

3.72 kB
Pairwise adversarial eval + Elo (the user's 'pairwise conditions') 3 months ago
playback.py

4.69 kB
Land PR #15 surgical fixes (Windows safety + human-play hints) 3 months ago
playback_view.py

9.78 kB
Quality drive: schema fix, 5 new/revised packs, 4 engine tests, scenario audit 3 months ago
playlist.py

14.1 kB
Add Playlist tab — cold-start non-gamer baseline UX 3 months ago
prompt_v2.py

16 kB
Add perception ablation grid (observation channel × fog of war) 3 months ago
providers.py

35.9 kB
providers: implement BedrockProvider via Converse API 3 months ago
resilience.py

7.25 kB
Eval resilience layer for real OpenRouter runs 3 months ago
run_eval.py

34.7 kB
run_eval: add --bedrock-region flag and route bedrock provider config 3 months ago
rust_adapter.py

18.3 kB
fix(engine): health: field on pre-placed actors + building actor ids surfaced for repair/sell 3 months ago
scoring.py

10.7 kB
win-speed bonus: record + reward how fast a model wins 3 months ago
structured_fog.py

3.25 kB
Structured-fog text mode, premium routing, codex descriptions, minimap colour-by-difficulty 3 months ago