#2429: added cerebras as a model provider.
docs
gateway
cli
commands
agents
Cluster:
Model Provider Integrations
# Add Cerebras Provider
## Summary
Adds Cerebras as a new model provider with support for 6 ultra-fast inference models using their OpenAI-compatible API at `https://api.cerebras.ai/v1`. Also they have a free tier of about 1M tokens a day, potentially allowing us to run clawdbot for free ( for some while ).
## Changes
- ✅ Added Cerebras provider configuration with 6 models:
- `llama3.1-8b` (default)
- `llama-3.3-70b`
- `gpt-oss-120b`
- `qwen-3-32b`
- `qwen-3-235b-a22b-instruct-2507`
- `zai-glm-4.7`
- ✅ Interactive onboarding support (`clawdbot onboard --auth-choice cerebras-api-key`)
- ✅ Non-interactive CLI flag (`--cerebras-api-key`)
- ✅ Documentation for setup and usage
## Files Modified
- `src/agents/models-config.providers.ts` - Provider configuration and model catalog
- `src/commands/onboard-auth.credentials.ts` - Credential management
- `src/commands/onboard-auth.config-core.ts` - Configuration helpers
- `src/commands/onboard-auth.ts` - Exported auth functions
- `src/commands/onboard-types.ts` - Type definitions
- `src/cli/program/register.onboard.ts` - CLI flag support
- `src/commands/auth-choice.apply.api-providers.ts` - Interactive auth handler
- `src/commands/onboard-non-interactive/local/auth-choice.ts` - Non-interactive auth handler
- `src/commands/auth-choice-options.ts` - Menu options and display
- `docs/providers/cerebras.md` - Provider documentation
- `docs/providers/index.md` - Provider index
- `docs/concepts/model-providers.md` - Model provider overview
## Testing
- [x] Build passes (`pnpm build`)
- [x] Lint passes (`pnpm lint`)
- [x] Tests pass (`pnpm test`)
- [x] Cerebras appears in interactive onboarding menu
- [x] CLI flag `--cerebras-api-key` works
- [x] Live API test with `CEREBRAS_API_KEY`
From the interactive onboarding form we can select cerebras as a model provider and use this awesome tool.
<!-- greptile_comment -->
<h2>Greptile Overview</h2>
<h3>Greptile Summary</h3>
Adds a new Cerebras model provider using an OpenAI-compatible API, including a model catalog, default model selection, and onboarding support (interactive and non-interactive) for storing a Cerebras API key and applying provider defaults.
This integrates with the existing provider-resolution flow (`resolveImplicitProviders`), onboarding config helpers (`apply*Config`), and the auth-choice selection system used by both interactive prompts and CLI flags.
<h3>Confidence Score: 2/5</h3>
- This PR is not safe to merge as-is due to onboarding regressions that can break auth selection and non-interactive configuration.
- Cerebras provider additions are straightforward, but `buildAuthChoiceOptions` appears to no longer populate choices (breaking interactive onboarding), and the new `--cerebras-api-key` flag is not wired through to `onboard()` so non-interactive onboarding won’t pick it up. There is also a breaking change removing `gateway-auth off` that may invalidate existing configs/scripts if not fully handled elsewhere.
- src/commands/auth-choice-options.ts, src/cli/program/register.onboard.ts, src/commands/onboard-types.ts
<!-- greptile_other_comments_section -->
<sub>(2/5) Greptile learns from your feedback when you react with thumbs up/down!</sub>
**Context used:**
- Context from `dashboard` - CLAUDE.md ([source](https://app.greptile.com/review/custom-context?memory=fd949e91-5c3a-4ab5-90a1-cbe184fd6ce8))
- Context from `dashboard` - AGENTS.md ([source](https://app.greptile.com/review/custom-context?memory=0d0c8278-ef8e-4d6c-ab21-f5527e322f13))
<!-- /greptile_comment -->
Most Similar PRs
#7051: Add io-intelligence model inference provider
by rajagurunath · 2026-02-02
81.9%
#11204: feat(providers): add Chutes (Bittensor) provider
by CodingAbdullah · 2026-02-07
80.8%
#7113: feat(providers): add CommonStack provider support
by flhoildy · 2026-02-02
80.1%
#7295: feat(provider) Add Key-less AI Integration via Puter
by ProgrammerIn-wonderland · 2026-02-02
80.0%
#21520: feat(onboard): add Cencori provider.
by bolaabanjo · 2026-02-20
77.9%
#9123: Feat/smart router backport and custom model provider
by JuliusYang3311 · 2026-02-04
76.7%
#13295: feat: add Eternal AI provider integration
by peterparkernho · 2026-02-10
76.7%
#15991: feat: add Novita AI provider support with dynamic model discovery
by Alex-wuhu · 2026-02-14
76.6%
#13079: feat: Add OpenAI-compatible API option to CLI for self-hosted models
by MikeWang0316tw · 2026-02-10
76.2%
#12020: feat: add AIsa provider, production-grade Chinese AI models
by renning22 · 2026-02-08
76.1%