MODELS
One list. Many providers.
The homepage table is the public catalog we publish. Context, latency, and token price there are marketing figures. Live routing and your own keys stay in the workspace.
| MODEL | CONTEXT | P50 LATENCY | / 1M TOKENS |
|---|---|---|---|
| LIVE gpt-6-astra | 128k | 0.8s | $4.20 |
| LIVE Opus 5.5 | 200k | 1.1s | $5.80 |
| LIVE Gemini 4 Pro | 1m | 0.9s | $3.60 |
| LIVE Qwen Ultra | 256k | 0.7s | $1.90 |