Every model Relay routes to, with current per-1M-token pricing. Relay automatically picks the cheapest capable model for each prompt — these are the ones it picks from.
Top-tier reasoning — used when correctness matters most.
| Model | Provider | Input / 1M | Output / 1M | Context |
|---|---|---|---|---|
| GPT-5 OpenAI's flagship — strongest general reasoning and multimodal. | OpenAI | $5.00 | $15.00 | 400K tokens |
| Claude Opus 4.7 Anthropic's flagship — top-tier coding and long-form writing. | Anthropic | $15.00 | $75.00 | 200K tokens |
| Gemini 2.5 Pro Massive context window and strong multimodal reasoning. | $1.25 | $10.00 | 2M tokens |
Workhorse models — Relay's default routing target.
| Model | Provider | Input / 1M | Output / 1M | Context |
|---|---|---|---|---|
| GPT-5 Mini Most of GPT-5's reasoning at a fraction of the price. | OpenAI | $0.40 | $1.60 | 400K tokens |
| Claude Sonnet 4.6 Sweet spot of Claude — strong coding at a fair price. | Anthropic | $3.00 | $15.00 | 200K tokens |
| Gemini 2.5 Flash Cheap, fast, multimodal — Relay's everyday default. | $0.30 | $2.50 | 1M tokens |
High-volume, low-cost — classification, drafts, simple Q&A.
| Model | Provider | Input / 1M | Output / 1M | Context |
|---|---|---|---|---|
| GPT-5 Nano Cheapest and fastest GPT-5 variant. | OpenAI | $0.10 | $0.40 | 400K tokens |
| Claude Haiku 4.5 Fast Claude for drafting and short answers. | Anthropic | $0.80 | $4.00 | 200K tokens |
| Gemini 2.0 Flash Among the cheapest production-grade LLMs available. | $0.10 | $0.40 | 1M tokens |
Relay routes each prompt to the cheapest model that can answer it well. Small teams typically cut their AI bill 40–70%.
Try Relay free →