II.
ProviderTranslation overview
Reference · liveprovider-translation:generic-openai:direct-providers
provider-translation:generic-openai:direct-providers overview
Inspect the raw attributes, linked wiki pages, and inbound or outbound graph edges for provider-translation:generic-openai:direct-providers.
Attributes
harness
generic-openai
provider
_group
providerGroup
- openai
- foundry
- groq
- fireworks
- together
- deepseek
- mistral
- cerebras
- sambanova
- openrouter
- xai
- grok
- ollama
- local
- lmstudio
- vllm
- custom
proxyRequired
false
envMapping
- envVarOPENAI_BASE_URLsourceparams.apiBaseconditionpresent
- envVarOPENAI_API_BASEsourceparams.apiBaseconditionpresent
- envVarOPENAI_API_KEYsourceauth.apiKeyconditionpresent
suppressNonAnthropicKey
true
args
[]
harnessUsers
- cursor
- omp
- openclaw
- hermes
- droid
- amp
notes
Generic OpenAI-compatible translation used by agents that speak the OpenAI protocol natively (Cursor, OMP, OpenClaw, Hermes, Droid, Amp). Sets both OPENAI_BASE_URL and OPENAI_API_BASE (legacy). Suppresses ANTHROPIC_API_KEY for non-anthropic providers. OpenClaw 2026.5.27 reuses xAI OAuth auth profiles for Grok web_search and carries the active agent auth profile through web_search calls; named auth profiles, OpenAI sampling params, Codex app-server resume/timeout/usage-limit recovery, OpenAI-compatible embedding providers in core, DeepInfra catalog browsing, Pixverse video generation, VLLM thinking params, Claude CLI OAuth overlays, direct Anthropic model IDs, xAI usage-limit surfacing, dynamic tool-schema guards, Ollama top-p normalization, and media-provider operation timeouts are tracked as provider recovery metadata.
Outgoing edges
None.
Incoming edges
None.