Apps · Roleplay & companions

Chub AI

Character hub with an integrated inference layer.

Tokens / 7d

64B

Routing directive

auto:chat

On our metal

35%

Licence

Proprietary

Chub routes long-running story sessions where context grows monotonically. Long-context routing keeps quality stable past 100K tokens instead of silently degrading.

→ chub.ai → All apps → Integration docs

How it connects

Custom endpoint with long-context routing enabled.
Context compaction handled client-side before the escalation threshold.
Content policy handled by the app, not the gateway.

Traffic pattern

Context grows linearly through a session, so per-message cost climbs unless compaction is applied.

deepseek-v4qwen3-maxllama-4-maverick

Point it at the gateway

{ "endpoint": "https://api.llmcloud.ai/v1",
  "model": "auto:long-context" }