Choosing a model

Model id shape

Vikey model ids are written provider/name, for example:

• anthropic/claude-sonnet-5.5
• openai/gpt-6.1-sol
• deepseek/deepseek-v4.1-flash
• moonshot/kimi-k3

The full list with prices is on the pricing page, or from the API:

curl https://api.vikey.ai/v1/models -H "Authorization: Bearer YOUR_API_KEY"

Short names

You may leave out the provider, for example claude-sonnet-5.5 or deepseek-v4.1-flash. Vikey looks for a model with that name:

• If exactly one matches, that model is used.
• If the name matches several providers, the request is rejected with 400 and the code ambiguous_model, listing the full ids to choose from. Vikey does not pick for you, because prices can differ.

For configs you keep for a long time, use the full id so it does not change when new models are added.

Hiding the provider name

API keys have an option to hide the provider name. When it is on, GET /v1/models shows short names (for example claude-sonnet-5.5) for models whose names do not clash. Requests with either the short name or the full id are still accepted.

Suggestions for coding

Need	Choice
You would rather not choose	Vikey Auto: vikey/agent-coding, vikey/agent-gpt or vikey/agent-mocin
Daily coding, high quality	anthropic/claude-sonnet-5.5 or openai/gpt-6.1-sol
Large refactors, hard bugs	anthropic/claude-opus-5.5
Many light tasks	deepseek/deepseek-v4.1-flash or openai/gpt-6-luna

Prices can change. The latest numbers are always on the pricing page and in the Models menu of the dashboard.