Model Catalog
Mock AI models available to the router, with cost, speed, sharpness, and reliability scores.
| Model | Provider | Description | Context | Input $/1M | Output $/1M | Speed | Sharpness | Reliability | Top skills |
|---|---|---|---|---|---|---|---|---|---|
| Balanced General Model | Nimbus AI | A well-rounded default model with solid performance across most everyday tasks. | 128,000 | $3.00 | $12.00 | 7/10 | 8/10 | 8/10 | codingwritingplanningresearchdebuggingreviewingtranslationbrainstormingsummarization |
| Cheap Fast Model | Nimbus AI | A low-cost, high-throughput model for simple, high-volume tasks. | 32,000 | $0.25 | $1.25 | 10/10 | 5/10 | 7/10 | |
| Coding Agent Model | Forge Systems | Specialized for agentic coding workflows: tool use, multi-file edits, and long coding sessions. | 128,000 | $5.00 | $20.00 | 6/10 | 9/10 | 8/10 | codingplanningdebuggingreviewingagentic-execution |
| Local Private Model | Self-hosted | A small self-hosted model for private, offline, zero-marginal-cost tasks. | 8,000 | $0.00 | $0.00 | 6/10 | 4/10 | 6/10 | |
| Long Context Model | Horizon AI | Built for huge context windows: document analysis and long conversation memory. | 1,000,000 | $4.00 | $16.00 | 5/10 | 8/10 | 8/10 | planningresearchreviewingsummarization |
| Premium Reasoning Model | Frontier AI Labs | A top-tier reasoning model for complex, high-stakes tasks where accuracy matters more than cost. | 200,000 | $15.00 | $75.00 | 5/10 | 10/10 | 9/10 | codingwritingplanningresearchdebuggingreviewingtranslationbrainstormingsummarizationagentic-execution |
| Writing Specialist Model | Wordcraft AI | Tuned for natural prose, tone control, and marketing copy. | 64,000 | $2.00 | $8.00 | 7/10 | 8/10 | 8/10 | writingbrainstormingsummarization |