3 models
Ranked by LiveCodeBench
Reasoning Tools JSON 93.2

Quality-first multi-agent model for hard research, analysis, and competitions

sakana/fugu-ultra 2026-06-15 1M context $5/M input $30/M output
11 providers
Reasoning Tools JSON 92.9

Multi-agent model for routing expert agents across complex analytical tasks

sakana/fugu 2026-06-15 1M context Input not listed Output not listed
1 provider
Reasoning Tools Open weights 89.0

Largest Nemotron 3 model for maximum open-weight reasoning and agent accuracy

nvidia/nemotron-3-ultra-550b-a55b 2026-06-04 1M context $0.5/M input $2.5/M output
15 providers