Quality-first multi-agent model for hard research, analysis, and competitions
14 providers
Model catalog
Models
Every model in the catalog with source-linked pricing, context limits, provider availability, and published benchmark results.
Multi-agent model for routing expert agents across complex analytical tasks
1 provider
Open-weight experimental preview of the Qwen4 architecture: hybrid-attention MoE (125B total, 6B active) with vision encoder for coding, agent tasks, and image and video understanding
6 providers
Dense 27B vision-language model for coding, agent tasks, and image and video understanding
44 providers
| Model | Creator | Score | Inputs | Context | Input | Output | Released | Compare |
|---|---|---|---|---|---|---|---|---|
| Fugu Ultrasakana/fugu-ultra | Sakana |
93.2 | 1M | $5 | $30 | 2026-06-15 | ||
| Fugusakana/fugu | Sakana |
92.9 | 1M | — | — | 2026-06-15 | ||
| Qwen3.8 Flash Nextalibaba/qwen3.8-flash-next | Alibaba Qwen |
91.9 | 262.144K | $0.1 | $0.2 | 2026-08-27 | ||
| Qwen3.8 27Balibaba/qwen3.8-27b | Alibaba Qwen |
90.3 | 262.144K | $0.04 | $1.35 | 2026-08-14 |