Added Gemini 4 Neon (Google), with three independent xhigh-effort dreams on the v1.1 corpus, to the leaderboard, Pareto chart, and task pages.
Models
Updated the v1.1 leaderboard, task-effort Pareto chart, and task pages with direct OpenAI results for GPT-6.2 Astra, GPT-6.2 Sol, and GPT-6 Sol. Each model used the Codex toolset for three independent max-effort dreams.
Models
Added Qwen4 Max (Alibaba) as a memory model, with three independent xhigh-effort dreams on the v1.1 corpus.
Models
Added Gemini 4 Hydrogen (Google) as a memory model, with three independent xhigh-effort dreams on the v1.1 corpus.
Models
Added GPT-6 Astra (OpenAI) as a memory model, with three independent runs on the v1.1 corpus.
Models
Added Claude Haiku 5.5 (Anthropic), GPT-6 Luna (OpenAI) and Muse Ember (Meta) as memory models, with three independent runs each on the v1.1 corpus.
Muse Wonder now uses memories dreamed at max effort (three runs on v1.1). The earlier v1.1 Muse Wonder memories were dreamed at xhigh.
Added Gemini 4 Carbon (Google) as a memory model, with three independent runs on the v1.1 corpus. Its memories were dreamed at high, Carbon's top effort.
Models
Added Claude Fable 5.2 (Anthropic) as a memory model, with three independent runs on the v1.1 corpus.
Models
Added Inkling 2 Small (Thinking Machines) as a memory model, with three independent runs on the v1.1 corpus.
Corpus
Updated the corpus to v1.1 (750 trajectories) to reduce reward hacking through trajectory metadata and balance out the corpus sources.
Results
Reran every model and harness on the v1.1 corpus, with three independent runs each.
Leaderboard scores are now the mean of the three runs, with ±1 standard deviation error bars.