fix: align controller with OpenShell and NemoClaw

Merged verified controller compatibility, terminal, and inference UX fixes.
This commit is contained in:
mmckeen-nv
2026-06-02 21:46:40 -06:00
committed by GitHub
parent 81f7239bf3
commit fafcf6e245
12 changed files with 224 additions and 87 deletions
+8 -2
View File
@@ -46,7 +46,7 @@ const providerTypeOptions = [
]
const OLLAMA_PROVIDER_NAME = "ollama-local"
const OLLAMA_BASE_URL = "http://host.docker.internal:11434/v1"
const OLLAMA_BASE_URL = "http://127.0.0.1:11434/v1"
const DEFAULT_VLLM_CONTAINER = "vllm-bigboy"
const DEFAULT_VLLM_IMAGE = "nvcr.io/nvidia/vllm:26.03.post1-py3"
const DEFAULT_VLLM_MODEL = "vllm-local"
@@ -467,6 +467,11 @@ export default function InferenceEndpointPanel() {
setProviders(Array.isArray(data.providers) ? data.providers : [])
if (data.gateway?.provider) setName(data.gateway.provider)
if (data.gateway?.model) setModel(data.gateway.model)
if (data.gateway?.provider?.toLowerCase().includes("ollama")) {
setType("openai")
setBaseUrl(OLLAMA_BASE_URL)
setCredentialKey("OLLAMA_API_KEY")
}
if (data.gateway?.provider && isVllmEndpoint(data.gateway.provider, "")) setType("vllm")
} catch (error) {
setMessage(error instanceof Error ? error.message : "Failed to load inference configuration")
@@ -516,7 +521,7 @@ export default function InferenceEndpointPanel() {
setType("openai")
setModel(modelName)
setBaseUrl(ollamaOpenAiBaseUrl(item))
setCredentialKey("OPENAI_API_KEY")
setCredentialKey("OLLAMA_API_KEY")
setApiKey("")
setRoute("gateway")
setNoVerify(true)
@@ -530,6 +535,7 @@ export default function InferenceEndpointPanel() {
setCredentialKey(provider.credentialKeys[0] || "OPENAI_API_KEY")
if (provider.name.toLowerCase().includes("ollama")) {
setBaseUrl(OLLAMA_BASE_URL)
setCredentialKey(provider.credentialKeys[0] || "OLLAMA_API_KEY")
setApiKey("")
}
setMessage(`Loaded ${provider.name}. Enter a model and save to make it active.`)