TokenReply

TokenReply is one of RoleCall's partner inference providers — an OpenAI-compatible provider that fronts a broad catalog of models behind a single API key. Connect it under Settings → Connections & Keys → BYOK Providers and its models appear in the Cast next to your other providers.

A trusted partner with free models. Sign up through tokenreply.com — you can try their free models before spending anything. The same link sits on the TokenReply card in the Add Provider screen.

---

Why TokenReply

• One OpenAI-compatible key. RoleCall points at https://api.tokenreply.com/v1 with your key, and every TokenReply model becomes pickable — streaming, samplers, and the model picker included.
• Top-ups and Plus. Pay as you go by topping up your balance, or subscribe to Plus for ongoing access.
• Broad catalog. Reach many vendors' models without juggling a separate key for each.

---

Connecting TokenReply in RoleCall

1. Open Settings → Connections & Keys → BYOK Providers → Add Provider and pick TokenReply.
2. Paste your API key — get one at tokenreply.com (free models available to try).
3. Optionally pin a Model ID (leave it blank to pick per-scene from the Cast).
4. Save, then hit Test Key to validate the key and cache the model list.

The full BYOK walkthrough — encryption modes, multiple configs, custom headers — lives in Providers, BYOK & Encryption and Setting Up BYOK Providers.

---

Samplers

TokenReply accepts the standard OpenAI-compatible sampler set:

Temperature	Top-P	Top-K	Frequency Penalty	Presence Penalty	Max Tokens
yes	yes	—	yes	yes	yes

For samplers not exposed natively (like min_p), drop them into Custom Body Params and they're merged into every request.

---

Tool calls

Native function-calling can be unreliable through providers that serve many vendors, TokenReply included. Leave the in-chat agent's tool-call mode on Auto — it defaults to Prompted, which is the more predictable path over routers. See Tool-Call Modes.

---

Good to know

• BYOK usage isn't metered by RoleCall — you pay TokenReply directly and their limits apply, not your RoleCall tier.
• Test your key after long breaks; provider catalogs and billing change, and Test Key re-primes the model list.