List your company

Reasoning on SynthWorlds 2026 (test)

0.48ScoreCodeAct
0.041120.155060.2690.38294May 12, 2026

Evaluation results

MethodLinks
CodeActBase Model=Llama-3.3 70B Instruct2026.050.48
RLMBase Model=Llama-3.3 70B Instruct2026.050.401
DOLORESBase Model=Llama-3.3 70B Instruct2026.050.359
DOLORESBase Model=Qwen3 32B Thinking2026.050.346
Deep ResearchBase Model=Llama-3.3 70B Instruct2026.050.308
DOLORESBase Model=Qwen3 8B Thinking2026.050.305
ReActBase Model=Llama-3.3 70B Instruct2026.050.3
CodeActBase Model=Qwen3 32B Thinking2026.050.281
Deep ResearchBase Model=Qwen3 32B Thinking2026.050.275
CodeActBase Model=Qwen3 8B Thinking2026.050.268
ReActBase Model=Qwen3 32B Thinking2026.050.228
Deep ResearchBase Model=Qwen3 8B Thinking2026.050.206
ReActBase Model=Qwen3 8B Thinking2026.050.176
RLMBase Model=Qwen3 32B Thinking2026.050.169
RLMBase Model=Qwen3 8B Thinking2026.050.058