Abliterated · Available
Abliterated Kimi K3.1 Pro API
Kimi K3.1 Pro. Available. Check the connected Router for current availability.
262,144 output tokens by default and maximum · 16 shared model slots per worker. 1,048,576-token shared context · 14,400-second configured request ceiling (up to 4 hours). Choose a smaller output budget in Chat or the API. Each worker's 16 model slots are shared across accounts. The active worker fleet varies. Accounts default to 32 concurrent requests across their keys, with operator overrides. A request can reach its deadline before exhausting its output budget.
Model facts
- Model ID
kimi-k31-pro-abliterated- Weights
- Abliterated
- Published rate snapshot
- $4.00 / $15.00USD per 1 million tokens · fresh input / outputCached input: $0.40 per 1M tokens
- Served precision
- See live Router metadata
- Access
- Verified account access. Live Router metadata supplies current readiness, supported features and immutable billing version.
- Status
- Available
The Router’s published price version governs billing.
Choose your model.
Check live availability in Chat or connect Codex and Hermes.
We work directly with model researchers and verify modified weights, commercial hosting rights, features and completed answers. Each served model’s identity stays explicit. We never substitute another model for your explicit choice.