dome models

dome models registers LLM upstreams the gateway proxies and groups them into routing pools that fan out across providers with priority, weighting, and failover.

Supported upstreams: Anthropic, OpenAI, Azure OpenAI, Google, Bedrock, hosted OpenAI-compatible vendors (Groq, Mistral, Together, Fireworks, DeepSeek, xAI, Perplexity, Cohere, OpenRouter, Cerebras, NVIDIA, DeepInfra, SambaNova, AI21, Databricks), and self-hosted endpoints.

Model rates

dome models rates [--model <substring>]

Show Dome's current model rate book and the byte-to-token conversion used when a provider returns no token usage. The default view requires authentication but no organization permission. It does not include the organization's negotiated provider rates.

dome models rates --effective
dome models rates --effective --model claude

--effective shows the rate card resolved for the active organization: its contract rates and explicit model associations layered over Dome's book. It requires the organization-scoped modelrates.manage permission, held by organization owners and admins by default.

Rates are micro-USD per one million tokens. Every rate is time-windowed: historical usage is priced at the era covering its own time rather than today's rate. Provider cost is an estimate of the organization's upstream bill, not a Dome charge. When token counts are unavailable, the displayed byte-to-token formula explains the additional estimate Dome applies.

Organization rate-card management is also available under this command:

Set an organization rate

dome models rates set <model-match> [flags]

Open a new pricing era. Supply literal rates or bind the model match to a Dome-owned rate with --priced-as.

Remove an organization rate

dome models rates remove <model-match>
dome models rates remove --era <era-id>

By default, removal closes the current era and retains it for historical pricing.

Edit a rate era

dome models rates edit-era <era-id> [flags]

Correct an existing era's rates, association, label, or boundaries. Omitted fields remain unchanged.

Show rate history

dome models rates history <model> [--org-id <org-id>]

Show the rate that wins for each historical time span and the source of that rate.

Suggest a rate association

dome models rates suggest <model>

Rank Dome-owned rate matches for a custom model. This command does not create an association.

Connections

List connections

dome models list [workspace-id-or-slug]

List LLM models in the workspace. Aliases: ls.

List providers

dome models providers

List the supported provider catalog and each provider's curated model ids. Use it to discover valid --provider and --model values for dome models add.

The catalog is backend-owned — the same source the dashboard Provider→Model picker and the MCP dome_models_providers tool read. Providers with stable public model ids (OpenAI, Anthropic, Google) carry a curated model list with one default. Other providers show an empty list and accept any model string the upstream accepts. The catalog is a convenience, not authoritative — a model your upstream supports but the catalog does not list is still valid.

dome models providers

dome models add claude-prod \
  --provider anthropic \
  --model "$(dome models providers --format json | jq -r '.providers[] | select(.provider=="anthropic") | .models[] | select(.default) | .id')" \
  --api-key "$ANTHROPIC_API_KEY"

Get connection

dome models get <name>

Show details for one model, including provider config and Cedar attributes.

Add connection

dome models add <name>

Register an LLM model upstream. The shared API key, when supplied, is stored under the provider's managed auth header — Authorization: Bearer … for OpenAI, Bedrock, and OpenAI-compatible hosted vendors; x-api-key for Anthropic; api-key for Azure OpenAI; x-goog-api-key for Google. When --endpoint is omitted, the CLI prefills the vendor's default base URL from the provider registry.

dome models add claude-sonnet \
  --provider anthropic \
  --model claude-3-5-sonnet-20241022 \
  --api-key "$ANTHROPIC_API_KEY"

dome models add fast-groq \
  --provider groq \
  --model llama-3.3-70b-versatile \
  --api-key "$GROQ_API_KEY"

dome models add gpt-4o \
  --provider azure_openai \
  --model gpt-4o \
  --endpoint https://my-resource.openai.azure.com \
  --api-key "$AZURE_OPENAI_KEY" \
  --provider-config '{"api_version":"2024-08-01-preview","deployment":"gpt-4o-prod"}'

Update connection

dome models update <name>

Update a model. Only provided flags change. Provider is immutable post-create — switch by adding a new connection.

Remove connection

dome models rm <name>

Remove an LLM model. Aliases: remove, delete.

OAuth

Connect OAuth

dome models oauth-connect <name>

Begin the shared-OAuth admin consent flow for a model configured with --auth-method=oauth --credential-type=shared: Anthropic OAuth, Azure AAD/Entra service principal, Google Workload Identity for Vertex. Prints a one-shot URL valid for ~10 minutes. Open it in a browser to complete consent — Dome stores the tokens on the callback.

Per-user OAuth models do not use this command; consent is gateway-triggered on each end user's first call.

Disconnect OAuth

dome models oauth-disconnect <name>

Revoke shared OAuth credentials. The OAuth client configuration is preserved, so a subsequent oauth-connect reuses the same client. Idempotent.

Pools

Create pool

dome models pool create <name>

Create a logical pool that fans out across one or more model connections.

dome models pool create default --failover-max 2
dome models pool member add default claude-sonnet --priority 0 --weight 3
dome models pool member add default gpt-4o --priority 0 --weight 1
dome models pool member add default claude-haiku --priority 1
dome models pool set-default default

# Drain the highest-headroom member first when members carry different spend caps
dome models pool create budget-aware \
  --routing-strategy most_quota_remaining

List pools

dome models pool list [workspace-id-or-slug]

List pools. Aliases: ls.

Get pool

dome models pool get <name>

Show pool config and its members.

Update pool

dome models pool update <name>

Update a pool. Only provided flags change. Pass --match-when '{}' to clear the match predicate.

Set default pool

dome models pool set-default <name>

Mark this pool as the workspace default. Replaces any prior default.

Move pool

dome models pool move <name> --before <anchor>
# or --after <anchor>

Reposition a pool in its workspace's match_when evaluation order. Pools with non-empty predicates evaluate in ascending sort_key; the first match wins. Refer to Pool order reference.

Exactly one of --before or --after is required.

Remove pool

dome models pool rm <name>

Remove a pool. Aliases: remove, delete.

Gateway membership

dome models pool gateways <list|add|remove> <pool-name-or-id> [gateway-name-or-id]

Manage the pool's membership in Gateways from the pool side. Mirrors dome gateways model-pools.

• list <pool> — Gateways this pool belongs to. Alias: ls.
• add <pool> <gateway> — expose the pool's models at the Gateway's /gateways/<id>/v1 endpoint.
• remove <pool> <gateway> — detach. Alias: rm.

Pool members

Add member

dome models pool member add <pool> <model>

Add a model to a pool. Failover walks ascending priority; within a bucket, requests distribute by weight.

List members

dome models pool member list <pool>

List members of a pool with priority, weight, and enabled state. Aliases: ls.

Update member

dome models pool member update <pool> <model>

Update a member's priority, weight, or enabled flag.

Remove member

dome models pool member rm <pool> <model>

Remove a member from a pool. Aliases: remove, delete.

Cap model spend, tokens, or calls with dome quotas. Use --subject model, --subject pool, --subject workspace, or --subject gateway as needed.