Models and providers
Pick a model, reasoning level, and fast mode, and connect your own API keys or ChatGPT, Claude, or Muse subscriptions.
Choose a model
Pick a model in the composer’s model menu. Models with reasoning levels have a reasoning submenu:
- None
- Low
- Medium
- High
- Extra High
- Max
Each model marks one level as its default. Models that support it also have a Fast switch, which uses the provider’s faster service tier. Sprocket saves the model, reasoning level, and fast mode for each thread.
Sprocket loads the model list from its servers, so new models show up without an update. To see the list from your terminal, run:
sprocket run --list-modelsProviders
A provider is where Sprocket sends model requests. Spikonado, Sprocket’s own provider, is on by default and uses your plan’s AI usage. The other providers use your own keys or subscriptions and don’t count against your Sprocket AI usage.
| Provider | How to connect |
|---|---|
| Spikonado | On by default |
| OpenAI API | API key |
| ChatGPT Subscription | Sign in with ChatGPT. You can add more than one account. Usage counts against your Codex allowance. |
| Claude Subscription | Sign in with Claude through Claude Code. Sprocket can install Claude Code for you. |
| Anthropic API | API key |
| Vercel AI Gateway | API key |
| OpenRouter | API key |
| OpenCode | API key |
| OpenCode Go | API key |
| Morph | API key. See Morph. |
| Muse Code Subscription | Sign in with Muse through the Muse CLI. Sprocket can install the Muse CLI for you. |
Add keys and subscriptions in Settings → BYOK/BYOS. Sprocket checks each API key with its provider before saving it.
Claude and Muse sign-in run on the computer that runs Sprocket. Open Sprocket on that computer to install them or sign in.
Morph
This is only in dev and canary builds. It isn’t in a stable release yet. See release channels.
Morph serves Kimi K3.1 Pro and DeepSeek V4.1 Flash 1017. Its Ultra plan has no token meter for both, but limits uncached input and output tokens per minute. When Morph rate-limits a request, Sprocket moves the run to the next enabled provider.
Provider order and fallback
In Settings → Providers, drag providers into the order you want and turn each on or off. At least one provider must stay on.
For each request, Sprocket uses the first enabled provider that offers the model. Sprocket retries server errors. If the provider still fails or rate-limits the request, Sprocket moves the run to the next enabled provider. If no enabled provider offers the model, the run fails with an error that points to Settings → Providers.
Long threads
When a thread’s context gets close to the model’s limit, the model writes a handoff note covering the task, its progress, and what’s left. Sprocket continues the thread in a fresh context that starts from that note. It does the same when a thread switches to another provider or to a model from a different vendor.