admin/openhuman
1
0
Fork 0
mirror of https://github.com/tinyhumansai/openhuman.git synced 2026-07-27 21:08:00 +00:00
Code Issues Packages Projects Releases Wiki Activity

Expand inference provider catalog (#3057) (#3062)

Browse Source
This commit is contained in:
Steven Enamakel
2026-05-30 22:37:06 -07:00
committed by GitHub
parent f64f10e3c1
commit 4e221e978b
12 changed files with 743 additions and 168 deletions
Download Patch File Download Diff File Expand all files Collapse all files
app/src/components/settings/panels/AIPanel.tsx
+26 -103
View File
@@ -53,6 +53,13 @@ import {
import { ConfirmationModal } from '../../intelligence/ConfirmationModal';
import SettingsHeader from '../components/SettingsHeader';
import { useSettingsNavigation } from '../hooks/useSettingsNavigation';
import {
authStyleForBuiltinCloudProvider,
BUILTIN_CLOUD_PROVIDER_META,
BUILTIN_CLOUD_PROVIDER_SLUGS,
builtinCloudProvider,
defaultEndpointForBuiltinCloudProvider,
} from './builtinCloudProviders';
import { presentProviderSetupError, ProviderSetupErrorNotice } from './ProviderSetupErrorNotice';
import { useReembedBackfillModal } from './useReembedBackfillModal';
@@ -105,6 +112,15 @@ const ROUTING_WORKLOAD_IDS: WorkloadId[] = [
'learning',
'subconscious',
];
const BUILTIN_RESERVED_SLUGS = [
'cloud',
'openhuman',
'pid',
'custom',
'ollama',
'lmstudio',
...BUILTIN_CLOUD_PROVIDER_SLUGS,
];
// ─────────────────────────────────────────────────────────────────────────────
// Static catalog
@@ -117,34 +133,7 @@ const BUILTIN_PROVIDER_META: Record<string, { tone: string; label: string }> = {
label: 'Managed',
tone: 'bg-emerald-50 dark:bg-emerald-500/10 ring-emerald-200 text-emerald-900 dark:text-emerald-100',
},
openai: {
label: 'OpenAI',
tone: 'bg-emerald-50 dark:bg-emerald-500/10 ring-emerald-200 text-emerald-900 dark:text-emerald-100',
},
anthropic: {
label: 'Anthropic',
tone: 'bg-orange-50 dark:bg-orange-500/10 ring-orange-200 text-orange-900 dark:text-orange-100',
},
openrouter: {
label: 'OpenRouter',
tone: 'bg-slate-100 dark:bg-slate-500/15 ring-slate-300 text-slate-900 dark:text-slate-100',
},
orcarouter: {
label: 'OrcaRouter',
tone: 'bg-sky-50 dark:bg-sky-500/10 ring-sky-200 text-sky-900 dark:text-sky-100',
},
gmi: {
label: 'GMI',
tone: 'bg-fuchsia-50 dark:bg-fuchsia-500/10 ring-fuchsia-200 text-fuchsia-900 dark:text-fuchsia-100',
},
fireworks: {
label: 'Fireworks',
tone: 'bg-rose-50 dark:bg-rose-500/10 ring-rose-200 text-rose-900 dark:text-rose-100',
},
moonshot: {
label: 'Kimi (Moonshot)',
tone: 'bg-indigo-50 dark:bg-indigo-500/10 ring-indigo-200 text-indigo-900 dark:text-indigo-100',
},
...BUILTIN_CLOUD_PROVIDER_META,
custom: {
label: 'Advanced',
tone: 'bg-sky-50 dark:bg-sky-500/10 ring-sky-200 text-sky-900 dark:text-sky-100',
@@ -259,9 +248,8 @@ function slugifyCustomProviderName(name: string): string {
*/
function authStyleForSlug(slug: string): AuthStyle {
if (slug === 'openhuman') return 'openhuman_jwt';
if (slug === 'anthropic') return 'anthropic';
if (slug === 'lmstudio' || slug === 'ollama') return 'none';
return 'bearer';
return authStyleForBuiltinCloudProvider(slug) ?? 'bearer';
}
function toPanelProvider(p: CloudProviderView): CloudProvider {
@@ -615,21 +603,7 @@ const ProviderKeyDialog = ({
const placeholder = isLocalRuntime
? defaultEndpointFor(slug) || t('settings.ai.defaultLocalEndpoint')
: slug === 'openai'
? 'sk-...'
: slug === 'anthropic'
? 'sk-ant-...'
: slug === 'openrouter'
? 'sk-or-...'
: slug === 'orcarouter'
? 'sk-orca-...'
: slug === 'gmi'
? 'gmi-...'
: slug === 'fireworks'
? 'fw-...'
: slug === 'moonshot'
? 'sk-...'
: 'your-api-key';
: (builtinCloudProvider(slug)?.keyPlaceholder ?? 'your-api-key');
const fieldLabel = isLocalRuntime
? t('settings.ai.endpointUrlLabel')
@@ -2808,18 +2782,8 @@ const AIPanel = ({ embedded = false }: AIPanelProps = {}) => {
onToggle={() => {}}
/>
{/* Built-in cloud providers — openai/anthropic/openrouter/orcarouter/custom */}
{(
[
'openai',
'anthropic',
'openrouter',
'orcarouter',
'gmi',
'fireworks',
'moonshot',
] as const
).map(slug => {
{/* Built-in cloud providers */}
{BUILTIN_CLOUD_PROVIDER_SLUGS.map(slug => {
const meta = BUILTIN_PROVIDER_META[slug];
const label = meta?.label ?? slug;
const existing = draft.cloudProviders.find(cp => cp.slug === slug);
@@ -2860,21 +2824,7 @@ const AIPanel = ({ embedded = false }: AIPanelProps = {}) => {
})}
{draft.cloudProviders
.filter(
cp =>
![
'openhuman',
'openai',
'anthropic',
'openrouter',
'orcarouter',
'gmi',
'fireworks',
'moonshot',
'lmstudio',
'ollama',
].includes(cp.slug)
)
.filter(cp => !BUILTIN_RESERVED_SLUGS.includes(cp.slug))
.map(existing => (
<ProviderToggleChip
key={existing.id}
@@ -3342,23 +3292,7 @@ const CloudProviderEditor = ({
const [saving, setSaving] = useState(false);
const [submitError, setSubmitError] = useState<string | null>(null);
const slug = initial?.slug ?? slugifyCustomProviderName(label);
const hasReservedSlugCollision =
!initial &&
[
'cloud',
'openhuman',
'pid',
'openai',
'anthropic',
'openrouter',
'orcarouter',
'gmi',
'fireworks',
'moonshot',
'custom',
'ollama',
'lmstudio',
].includes(slug);
const hasReservedSlugCollision = !initial && BUILTIN_RESERVED_SLUGS.includes(slug);
const slugError = !slug
? t('settings.ai.slugMissingError')
: existingSlugs.includes(slug)
@@ -3504,23 +3438,12 @@ const CloudProviderEditor = ({
};
function defaultEndpointFor(slug: string): string {
const builtinEndpoint = defaultEndpointForBuiltinCloudProvider(slug);
if (builtinEndpoint) return builtinEndpoint;
switch (slug) {
case 'openhuman':
return 'https://api.openhuman.ai/v1';
case 'openai':
return 'https://api.openai.com/v1';
case 'anthropic':
return 'https://api.anthropic.com/v1';
case 'openrouter':
return 'https://openrouter.ai/api/v1';
case 'orcarouter':
return 'https://api.orcarouter.ai/v1';
case 'gmi':
return 'https://api.gmi-serving.com/v1';
case 'fireworks':
return 'https://api.fireworks.ai/inference/v1';
case 'moonshot':
return 'https://api.moonshot.ai/v1';
case 'ollama':
// Ollama exposes an OpenAI-compatible endpoint at /v1; the bare host is
// also accepted by the Rust factory (it appends /v1 internally for chat).
app/src/components/settings/panels/__tests__/AIPanel.test.tsx
+76
View File
@@ -380,6 +380,82 @@ describe('AIPanel', () => {
).not.toBeInTheDocument();
});
it('renders Phase 1 built-in provider chips including SumoPod', async () => {
vi.mocked(loadAISettings).mockResolvedValue({ ...baseSettings, cloudProviders: [] });
renderWithProviders(<AIPanel />);
for (const label of ['Groq', 'DeepSeek', 'MiniMax', 'SumoPod']) {
await waitFor(() =>
expect(
screen.getByRole('switch', { name: new RegExp(`Connect ${label}`, 'i') })
).toBeInTheDocument()
);
}
});
it('connects SumoPod with the native endpoint and provider:sumopod key', async () => {
vi.mocked(loadAISettings).mockResolvedValue({ ...baseSettings, cloudProviders: [] });
renderWithProviders(<AIPanel />);
fireEvent.click(await screen.findByRole('switch', { name: /Connect SumoPod/i }));
const dialog = await screen.findByRole('dialog', { name: /Connect SumoPod/i });
fireEvent.change(within(dialog).getByLabelText(/API key/i), {
target: { value: 'sk-sumopod-test' },
});
fireEvent.click(within(dialog).getByRole('button', { name: /^Save$/i }));
await waitFor(() =>
expect(vi.mocked(setCloudProviderKey)).toHaveBeenCalledWith('sumopod', 'sk-sumopod-test')
);
await waitFor(() => expect(vi.mocked(listProviderModels)).toHaveBeenCalledWith('sumopod'));
await waitFor(() => expect(vi.mocked(saveAISettings)).toHaveBeenCalled());
const [, nextSettings] = vi.mocked(saveAISettings).mock.calls[0];
expect(nextSettings.cloudProviders).toEqual(
expect.arrayContaining([
expect.objectContaining({
slug: 'sumopod',
label: 'SumoPod',
endpoint: 'https://ai.sumopod.com/v1',
auth_style: 'bearer',
has_api_key: true,
}),
])
);
});
it('connects MiniMax with anthropic auth style', async () => {
vi.mocked(loadAISettings).mockResolvedValue({ ...baseSettings, cloudProviders: [] });
renderWithProviders(<AIPanel />);
fireEvent.click(await screen.findByRole('switch', { name: /Connect MiniMax/i }));
const dialog = await screen.findByRole('dialog', { name: /Connect MiniMax/i });
fireEvent.change(within(dialog).getByLabelText(/API key/i), {
target: { value: 'sk-minimax-test' },
});
fireEvent.click(within(dialog).getByRole('button', { name: /^Save$/i }));
await waitFor(() =>
expect(vi.mocked(setCloudProviderKey)).toHaveBeenCalledWith('minimax', 'sk-minimax-test')
);
await waitFor(() => expect(vi.mocked(saveAISettings)).toHaveBeenCalled());
const [, nextSettings] = vi.mocked(saveAISettings).mock.calls[0];
expect(nextSettings.cloudProviders).toEqual(
expect.arrayContaining([
expect.objectContaining({
slug: 'minimax',
label: 'MiniMax',
endpoint: 'https://api.minimax.io/anthropic',
auth_style: 'anthropic',
}),
])
);
});
it('surfaces provider setup errors in an alert with technical details collapsed', async () => {
vi.mocked(loadAISettings).mockResolvedValue({ ...baseSettings, cloudProviders: [] });
vi.mocked(listProviderModels).mockRejectedValueOnce(
app/src/components/settings/panels/__tests__/builtinCloudProviders.test.ts
+49
View File
@@ -0,0 +1,49 @@
import { describe, expect, it } from 'vitest';
import {
authStyleForBuiltinCloudProvider,
BUILTIN_CLOUD_PROVIDER_SLUGS,
BUILTIN_CLOUD_PROVIDERS,
defaultEndpointForBuiltinCloudProvider,
} from '../builtinCloudProviders';
describe('builtinCloudProviders', () => {
it('keeps built-in provider slugs unique', () => {
expect(new Set(BUILTIN_CLOUD_PROVIDER_SLUGS).size).toBe(BUILTIN_CLOUD_PROVIDER_SLUGS.length);
});
it.each([
['groq', 'https://api.groq.com/openai/v1', 'bearer'],
['deepseek', 'https://api.deepseek.com/v1', 'bearer'],
['minimax', 'https://api.minimax.io/anthropic', 'anthropic'],
['sumopod', 'https://ai.sumopod.com/v1', 'bearer'],
] as const)('maps %s to its endpoint and auth style', (slug, endpoint, authStyle) => {
expect(defaultEndpointForBuiltinCloudProvider(slug)).toBe(endpoint);
expect(authStyleForBuiltinCloudProvider(slug)).toBe(authStyle);
});
it('contains the full phase one provider set', () => {
expect(BUILTIN_CLOUD_PROVIDERS.map(provider => provider.slug)).toEqual(
expect.arrayContaining([
'groq',
'mistral',
'deepseek',
'together',
'google',
'cerebras',
'xai',
'huggingface',
'nvidia',
'zai',
'minimax',
'stepfun',
'kilocode',
'deepinfra',
'novita',
'venice',
'vercel-ai-gateway',
'sumopod',
])
);
});
});
app/src/components/settings/panels/builtinCloudProviders.ts
+236
View File
@@ -0,0 +1,236 @@
import type { AuthStyle } from '../../../utils/tauriCommands/config';
export type BuiltinCloudProvider = {
slug: string;
label: string;
endpoint: string;
authStyle: AuthStyle;
tone: string;
keyPlaceholder?: string;
};
const TONE = {
emerald:
'bg-emerald-50 dark:bg-emerald-500/10 ring-emerald-200 text-emerald-900 dark:text-emerald-100',
orange: 'bg-orange-50 dark:bg-orange-500/10 ring-orange-200 text-orange-900 dark:text-orange-100',
slate: 'bg-slate-100 dark:bg-slate-500/15 ring-slate-300 text-slate-900 dark:text-slate-100',
sky: 'bg-sky-50 dark:bg-sky-500/10 ring-sky-200 text-sky-900 dark:text-sky-100',
fuchsia:
'bg-fuchsia-50 dark:bg-fuchsia-500/10 ring-fuchsia-200 text-fuchsia-900 dark:text-fuchsia-100',
rose: 'bg-rose-50 dark:bg-rose-500/10 ring-rose-200 text-rose-900 dark:text-rose-100',
indigo: 'bg-indigo-50 dark:bg-indigo-500/10 ring-indigo-200 text-indigo-900 dark:text-indigo-100',
amber: 'bg-amber-50 dark:bg-amber-500/10 ring-amber-200 text-amber-900 dark:text-amber-100',
teal: 'bg-teal-50 dark:bg-teal-500/10 ring-teal-200 text-teal-900 dark:text-teal-100',
violet: 'bg-violet-50 dark:bg-violet-500/10 ring-violet-200 text-violet-900 dark:text-violet-100',
zinc: 'bg-zinc-100 dark:bg-zinc-500/15 ring-zinc-300 text-zinc-900 dark:text-zinc-100',
} as const;
export const BUILTIN_CLOUD_PROVIDERS: BuiltinCloudProvider[] = [
{
slug: 'openai',
label: 'OpenAI',
endpoint: 'https://api.openai.com/v1',
authStyle: 'bearer',
tone: TONE.emerald,
keyPlaceholder: 'sk-...',
},
{
slug: 'anthropic',
label: 'Anthropic',
endpoint: 'https://api.anthropic.com/v1',
authStyle: 'anthropic',
tone: TONE.orange,
keyPlaceholder: 'sk-ant-...',
},
{
slug: 'openrouter',
label: 'OpenRouter',
endpoint: 'https://openrouter.ai/api/v1',
authStyle: 'bearer',
tone: TONE.slate,
keyPlaceholder: 'sk-or-...',
},
{
slug: 'orcarouter',
label: 'OrcaRouter',
endpoint: 'https://api.orcarouter.ai/v1',
authStyle: 'bearer',
tone: TONE.sky,
keyPlaceholder: 'sk-orca-...',
},
{
slug: 'gmi',
label: 'GMI',
endpoint: 'https://api.gmi-serving.com/v1',
authStyle: 'bearer',
tone: TONE.fuchsia,
keyPlaceholder: 'gmi-...',
},
{
slug: 'fireworks',
label: 'Fireworks',
endpoint: 'https://api.fireworks.ai/inference/v1',
authStyle: 'bearer',
tone: TONE.rose,
keyPlaceholder: 'fw-...',
},
{
slug: 'moonshot',
label: 'Kimi (Moonshot)',
endpoint: 'https://api.moonshot.ai/v1',
authStyle: 'bearer',
tone: TONE.indigo,
keyPlaceholder: 'sk-...',
},
{
slug: 'groq',
label: 'Groq',
endpoint: 'https://api.groq.com/openai/v1',
authStyle: 'bearer',
tone: TONE.teal,
keyPlaceholder: 'gsk_...',
},
{
slug: 'mistral',
label: 'Mistral',
endpoint: 'https://api.mistral.ai/v1',
authStyle: 'bearer',
tone: TONE.amber,
},
{
slug: 'deepseek',
label: 'DeepSeek',
endpoint: 'https://api.deepseek.com/v1',
authStyle: 'bearer',
tone: TONE.zinc,
keyPlaceholder: 'sk-...',
},
{
slug: 'together',
label: 'Together AI',
endpoint: 'https://api.together.xyz/v1',
authStyle: 'bearer',
tone: TONE.violet,
},
{
slug: 'google',
label: 'Google Gemini',
endpoint: 'https://generativelanguage.googleapis.com/v1beta/openai',
authStyle: 'bearer',
tone: TONE.sky,
},
{
slug: 'cerebras',
label: 'Cerebras',
endpoint: 'https://api.cerebras.ai/v1',
authStyle: 'bearer',
tone: TONE.orange,
},
{
slug: 'xai',
label: 'xAI',
endpoint: 'https://api.x.ai/v1',
authStyle: 'bearer',
tone: TONE.zinc,
},
{
slug: 'huggingface',
label: 'Hugging Face',
endpoint: 'https://router.huggingface.co/v1',
authStyle: 'bearer',
tone: TONE.amber,
keyPlaceholder: 'hf_...',
},
{
slug: 'nvidia',
label: 'NVIDIA',
endpoint: 'https://integrate.api.nvidia.com/v1',
authStyle: 'bearer',
tone: TONE.emerald,
},
{
slug: 'zai',
label: 'Z.AI',
endpoint: 'https://api.z.ai/api/paas/v4',
authStyle: 'bearer',
tone: TONE.teal,
},
{
slug: 'minimax',
label: 'MiniMax',
endpoint: 'https://api.minimax.io/anthropic',
authStyle: 'anthropic',
tone: TONE.rose,
},
{
slug: 'stepfun',
label: 'StepFun',
endpoint: 'https://api.stepfun.ai/step_plan/v1',
authStyle: 'bearer',
tone: TONE.indigo,
},
{
slug: 'kilocode',
label: 'Kilo Code',
endpoint: 'https://api.kilo.ai/api/gateway',
authStyle: 'bearer',
tone: TONE.fuchsia,
},
{
slug: 'deepinfra',
label: 'DeepInfra',
endpoint: 'https://api.deepinfra.com/v1/openai',
authStyle: 'bearer',
tone: TONE.slate,
},
{
slug: 'novita',
label: 'Novita',
endpoint: 'https://api.novita.ai/v3/openai',
authStyle: 'bearer',
tone: TONE.violet,
},
{
slug: 'venice',
label: 'Venice',
endpoint: 'https://api.venice.ai/api/v1',
authStyle: 'bearer',
tone: TONE.teal,
},
{
slug: 'vercel-ai-gateway',
label: 'Vercel AI Gateway',
endpoint: 'https://ai-gateway.vercel.sh/v1',
authStyle: 'bearer',
tone: TONE.zinc,
},
{
slug: 'sumopod',
label: 'SumoPod',
endpoint: 'https://ai.sumopod.com/v1',
authStyle: 'bearer',
tone: TONE.amber,
keyPlaceholder: 'sk-...',
},
];
export const BUILTIN_CLOUD_PROVIDER_SLUGS = BUILTIN_CLOUD_PROVIDERS.map(provider => provider.slug);
export const BUILTIN_CLOUD_PROVIDER_META = Object.fromEntries(
BUILTIN_CLOUD_PROVIDERS.map(provider => [
provider.slug,
{ label: provider.label, tone: provider.tone },
])
) as Record<string, { tone: string; label: string }>;
export function builtinCloudProvider(slug: string): BuiltinCloudProvider | undefined {
return BUILTIN_CLOUD_PROVIDERS.find(provider => provider.slug === slug);
}
export function defaultEndpointForBuiltinCloudProvider(slug: string): string {
return builtinCloudProvider(slug)?.endpoint ?? '';
}
export function authStyleForBuiltinCloudProvider(slug: string): AuthStyle | undefined {
return builtinCloudProvider(slug)?.authStyle;
}
docs/inference-provider-catalog.md
+52
View File
@@ -0,0 +1,52 @@
# Inference Provider Catalog
This document tracks the built-in BYOK inference-provider presets exposed in
Settings > AI. OpenHuman still supports arbitrary OpenAI-compatible endpoints
through a custom provider entry; this table is the first-class chip catalog.
## Phase 1 Presets
| Slug | Label | Endpoint | Auth style | Status |
| ------------------- | ----------------- | --------------------------------------------------------- | ---------- | ------- |
| `openai` | OpenAI | `https://api.openai.com/v1` | bearer | Shipped |
| `anthropic` | Anthropic | `https://api.anthropic.com/v1` | anthropic | Shipped |
| `openrouter` | OpenRouter | `https://openrouter.ai/api/v1` | bearer | Shipped |
| `orcarouter` | OrcaRouter | `https://api.orcarouter.ai/v1` | bearer | Shipped |
| `gmi` | GMI | `https://api.gmi-serving.com/v1` | bearer | Shipped |
| `fireworks` | Fireworks | `https://api.fireworks.ai/inference/v1` | bearer | Shipped |
| `moonshot` | Kimi (Moonshot) | `https://api.moonshot.ai/v1` | bearer | Shipped |
| `groq` | Groq | `https://api.groq.com/openai/v1` | bearer | Shipped |
| `mistral` | Mistral | `https://api.mistral.ai/v1` | bearer | Shipped |
| `deepseek` | DeepSeek | `https://api.deepseek.com/v1` | bearer | Shipped |
| `together` | Together AI | `https://api.together.xyz/v1` | bearer | Shipped |
| `google` | Google Gemini | `https://generativelanguage.googleapis.com/v1beta/openai` | bearer | Shipped |
| `cerebras` | Cerebras | `https://api.cerebras.ai/v1` | bearer | Shipped |
| `xai` | xAI | `https://api.x.ai/v1` | bearer | Shipped |
| `huggingface` | Hugging Face | `https://router.huggingface.co/v1` | bearer | Shipped |
| `nvidia` | NVIDIA | `https://integrate.api.nvidia.com/v1` | bearer | Shipped |
| `zai` | Z.AI | `https://api.z.ai/api/paas/v4` | bearer | Shipped |
| `minimax` | MiniMax | `https://api.minimax.io/anthropic` | anthropic | Shipped |
| `stepfun` | StepFun | `https://api.stepfun.ai/step_plan/v1` | bearer | Shipped |
| `kilocode` | Kilo Code | `https://api.kilo.ai/api/gateway` | bearer | Shipped |
| `deepinfra` | DeepInfra | `https://api.deepinfra.com/v1/openai` | bearer | Shipped |
| `novita` | Novita | `https://api.novita.ai/v3/openai` | bearer | Shipped |
| `venice` | Venice | `https://api.venice.ai/api/v1` | bearer | Shipped |
| `vercel-ai-gateway` | Vercel AI Gateway | `https://ai-gateway.vercel.sh/v1` | bearer | Shipped |
| `sumopod` | SumoPod | `https://ai.sumopod.com/v1` | bearer | Shipped |
API keys are stored through the auth-profile store under `provider:<slug>`;
they are not read from environment variables by the desktop Settings flow.
## Deferred Transports
The following provider families need transport or credential work beyond a
static OpenAI-compatible or Anthropic-compatible preset:
| Provider / family | Needed work |
| ------------------------------------------------------ | ---------------------------------------------------------------------- |
| OpenAI Codex / ChatGPT subscription | Continue extending existing `openai_oauth` support. |
| xAI OAuth, MiniMax OAuth, Qwen OAuth, Nous device-code | Add provider-specific OAuth or import flows. |
| Google Gemini CLI, Claude CLI, Copilot ACP | Add subprocess or external-process transports. |
| AWS Bedrock Converse | Add native AWS SDK / Bedrock Converse transport. |
| Azure Foundry | Model dual OpenAI / Anthropic API modes and Azure credential handling. |
| GitHub Copilot | Add GitHub-token import and provider transport. |
src/openhuman/about_app/catalog_data.rs
+2 -2
View File
@@ -1087,8 +1087,8 @@ pub(super) const CAPABILITIES: &[Capability] = &[
name: "Configure AI",
domain: "settings",
category: CapabilityCategory::Settings,
description: "Adjust AI-related settings and agent behavior preferences.",
how_to: "Settings > Developer Options > AI Configuration",
description: "Configure managed, local, custom, and built-in BYOK LLM providers, including SumoPod and other OpenAI-compatible gateways, plus per-workload routing preferences.",
how_to: "Settings > AI",
status: CapabilityStatus::Stable,
privacy: None,
},
src/openhuman/config/schema/cloud_providers.rs
+239 -26
View File
@@ -12,6 +12,179 @@
use schemars::JsonSchema;
use serde::{Deserialize, Serialize};
#[derive(Debug, Clone, Copy, PartialEq, Eq)]
pub struct BuiltinCloudProvider {
pub slug: &'static str,
pub label: &'static str,
pub endpoint: &'static str,
pub auth_style: AuthStyle,
}
pub const BUILTIN_CLOUD_PROVIDERS: &[BuiltinCloudProvider] = &[
BuiltinCloudProvider {
slug: "openhuman",
label: "OpenHuman",
endpoint: "https://api.openhuman.ai/v1",
auth_style: AuthStyle::OpenhumanJwt,
},
BuiltinCloudProvider {
slug: "openai",
label: "OpenAI",
endpoint: "https://api.openai.com/v1",
auth_style: AuthStyle::Bearer,
},
BuiltinCloudProvider {
slug: "anthropic",
label: "Anthropic",
endpoint: "https://api.anthropic.com/v1",
auth_style: AuthStyle::Anthropic,
},
BuiltinCloudProvider {
slug: "openrouter",
label: "OpenRouter",
endpoint: "https://openrouter.ai/api/v1",
auth_style: AuthStyle::Bearer,
},
BuiltinCloudProvider {
slug: "orcarouter",
label: "OrcaRouter",
endpoint: "https://api.orcarouter.ai/v1",
auth_style: AuthStyle::Bearer,
},
BuiltinCloudProvider {
slug: "gmi",
label: "GMI",
endpoint: "https://api.gmi-serving.com/v1",
auth_style: AuthStyle::Bearer,
},
BuiltinCloudProvider {
slug: "fireworks",
label: "Fireworks",
endpoint: "https://api.fireworks.ai/inference/v1",
auth_style: AuthStyle::Bearer,
},
BuiltinCloudProvider {
slug: "moonshot",
label: "Kimi (Moonshot)",
endpoint: "https://api.moonshot.ai/v1",
auth_style: AuthStyle::Bearer,
},
BuiltinCloudProvider {
slug: "groq",
label: "Groq",
endpoint: "https://api.groq.com/openai/v1",
auth_style: AuthStyle::Bearer,
},
BuiltinCloudProvider {
slug: "mistral",
label: "Mistral",
endpoint: "https://api.mistral.ai/v1",
auth_style: AuthStyle::Bearer,
},
BuiltinCloudProvider {
slug: "deepseek",
label: "DeepSeek",
endpoint: "https://api.deepseek.com/v1",
auth_style: AuthStyle::Bearer,
},
BuiltinCloudProvider {
slug: "together",
label: "Together AI",
endpoint: "https://api.together.xyz/v1",
auth_style: AuthStyle::Bearer,
},
BuiltinCloudProvider {
slug: "google",
label: "Google Gemini",
endpoint: "https://generativelanguage.googleapis.com/v1beta/openai",
auth_style: AuthStyle::Bearer,
},
BuiltinCloudProvider {
slug: "cerebras",
label: "Cerebras",
endpoint: "https://api.cerebras.ai/v1",
auth_style: AuthStyle::Bearer,
},
BuiltinCloudProvider {
slug: "xai",
label: "xAI",
endpoint: "https://api.x.ai/v1",
auth_style: AuthStyle::Bearer,
},
BuiltinCloudProvider {
slug: "huggingface",
label: "Hugging Face",
endpoint: "https://router.huggingface.co/v1",
auth_style: AuthStyle::Bearer,
},
BuiltinCloudProvider {
slug: "nvidia",
label: "NVIDIA",
endpoint: "https://integrate.api.nvidia.com/v1",
auth_style: AuthStyle::Bearer,
},
BuiltinCloudProvider {
slug: "zai",
label: "Z.AI",
endpoint: "https://api.z.ai/api/paas/v4",
auth_style: AuthStyle::Bearer,
},
BuiltinCloudProvider {
slug: "minimax",
label: "MiniMax",
endpoint: "https://api.minimax.io/anthropic",
auth_style: AuthStyle::Anthropic,
},
BuiltinCloudProvider {
slug: "stepfun",
label: "StepFun",
endpoint: "https://api.stepfun.ai/step_plan/v1",
auth_style: AuthStyle::Bearer,
},
BuiltinCloudProvider {
slug: "kilocode",
label: "Kilo Code",
endpoint: "https://api.kilo.ai/api/gateway",
auth_style: AuthStyle::Bearer,
},
BuiltinCloudProvider {
slug: "deepinfra",
label: "DeepInfra",
endpoint: "https://api.deepinfra.com/v1/openai",
auth_style: AuthStyle::Bearer,
},
BuiltinCloudProvider {
slug: "novita",
label: "Novita",
endpoint: "https://api.novita.ai/v3/openai",
auth_style: AuthStyle::Bearer,
},
BuiltinCloudProvider {
slug: "venice",
label: "Venice",
endpoint: "https://api.venice.ai/api/v1",
auth_style: AuthStyle::Bearer,
},
BuiltinCloudProvider {
slug: "vercel-ai-gateway",
label: "Vercel AI Gateway",
endpoint: "https://ai-gateway.vercel.sh/v1",
auth_style: AuthStyle::Bearer,
},
BuiltinCloudProvider {
slug: "sumopod",
label: "SumoPod",
endpoint: "https://ai.sumopod.com/v1",
auth_style: AuthStyle::Bearer,
},
];
fn builtin_cloud_provider(type_str: &str) -> Option<&'static BuiltinCloudProvider> {
BUILTIN_CLOUD_PROVIDERS
.iter()
.find(|provider| provider.slug == type_str)
}
/// Authentication header style for a cloud provider.
///
/// Wire format is lowercase (e.g. `"bearer"`). Determines which HTTP headers
@@ -160,41 +333,24 @@ pub fn migrate_legacy_fields(entry: &mut CloudProviderCreds) {
// Auth style from legacy type when still at default Bearer.
if entry.auth_style == AuthStyle::Bearer {
match lt {
"anthropic" => {
entry.auth_style = AuthStyle::Anthropic;
}
"openhuman" => {
entry.auth_style = AuthStyle::OpenhumanJwt;
}
_ => {}
if let Some(provider) = builtin_cloud_provider(lt) {
entry.auth_style = provider.auth_style;
}
}
}
/// Map a legacy type string (or slug) to a human-readable label.
fn legacy_label_for(type_str: &str) -> &'static str {
match type_str {
"openhuman" => "OpenHuman",
"openai" => "OpenAI",
"anthropic" => "Anthropic",
"openrouter" => "OpenRouter",
"orcarouter" => "OrcaRouter",
"custom" => "Custom",
_ => "Custom",
}
builtin_cloud_provider(type_str)
.map(|provider| provider.label)
.unwrap_or("Custom")
}
/// Map a legacy type string to its well-known default endpoint.
fn legacy_default_endpoint(type_str: &str) -> &'static str {
match type_str {
"openhuman" => "https://api.openhuman.ai/v1",
"openai" => "https://api.openai.com/v1",
"anthropic" => "https://api.anthropic.com/v1",
"openrouter" => "https://openrouter.ai/api/v1",
"orcarouter" => "https://api.orcarouter.ai/v1",
_ => "",
}
builtin_cloud_provider(type_str)
.map(|provider| provider.endpoint)
.unwrap_or("")
}
/// Generate a short opaque id for a new provider entry.
@@ -303,7 +459,10 @@ impl CloudProviderType {
#[cfg(test)]
mod tests {
use super::is_slug_reserved;
use super::{
is_slug_reserved, migrate_legacy_fields, AuthStyle, CloudProviderCreds,
BUILTIN_CLOUD_PROVIDERS,
};
#[test]
fn reserved_slugs() {
@@ -329,4 +488,58 @@ mod tests {
"lmstudio is a free-form OpenAI-compatible slug"
);
}
#[test]
fn builtin_cloud_provider_defaults_cover_phase_one_presets() {
for (slug, label, endpoint, auth_style) in [
(
"groq",
"Groq",
"https://api.groq.com/openai/v1",
AuthStyle::Bearer,
),
(
"deepseek",
"DeepSeek",
"https://api.deepseek.com/v1",
AuthStyle::Bearer,
),
(
"minimax",
"MiniMax",
"https://api.minimax.io/anthropic",
AuthStyle::Anthropic,
),
(
"sumopod",
"SumoPod",
"https://ai.sumopod.com/v1",
AuthStyle::Bearer,
),
] {
let mut entry = CloudProviderCreds {
id: format!("p_{slug}"),
legacy_type: Some(slug.to_string()),
..Default::default()
};
migrate_legacy_fields(&mut entry);
assert_eq!(entry.slug, slug);
assert_eq!(entry.label, label);
assert_eq!(entry.endpoint, endpoint);
assert_eq!(entry.auth_style, auth_style);
}
}
#[test]
fn builtin_cloud_provider_slugs_are_unique() {
let mut slugs = std::collections::HashSet::new();
for provider in BUILTIN_CLOUD_PROVIDERS {
assert!(
slugs.insert(provider.slug),
"duplicate built-in cloud provider slug {}",
provider.slug
);
}
}
}
src/openhuman/inference/README.md
+40 -37
View File
@@ -12,46 +12,49 @@ Unified inference domain: the canonical home for everything LLM/STT/TTS/embeddin
- Run ChatGPT/Codex OAuth (PKCE) for the `openai` cloud slug and persist tokens in the encrypted auth-profile store.
- Expose an OpenAI-compatible `/v1/*` HTTP endpoint guarded by a stable user-managed external bearer.
- Detect device hardware profile and recommend/apply local model presets/tiers.
- Maintain the built-in BYOK provider preset catalog used by Settings > AI.
The current matrix lives in `docs/inference-provider-catalog.md`; credentials
are stored under `provider:<slug>` in the auth-profile store.
## Key files
| File / dir | Role |
| --- | --- |
| `mod.rs` | Domain root; module decls + re-exports; wires `inference.*` controller schemas/controllers. |
| `ops.rs` | Canonical handler file — `inference_*` business logic returning `RpcOutcome<T>`; delegates to `local`, `provider`, `sentiment`, `device`, `presets`, `openai_oauth`. Includes Sentry-noise suppression for expected provider/user-config failures. |
| `schemas.rs` | `inference.*` controller schemas + `handle_*` fns + param DTOs. |
| `types.rs` | Serde DTOs: `LocalAiStatus`, `LocalAiAssetsStatus`, `LocalAiDownloadsProgress`, `LocalAiEmbeddingResult`, `LocalAiSpeechResult`, `LocalAiTtsResult`, etc. |
| `device.rs` | `DeviceProfile` hardware detection (RAM/CPU/GPU/OS), cached. |
| `model_ids.rs` | Effective chat/vision/embedding/STT/TTS/quantization model id resolution from config. |
| `model_context.rs` | Known model context-window sizes (`context_window_for_model`) for pre-dispatch budgeting. |
| `presets.rs` | `ModelPreset`, `ModelTier`, `VisionMode`; tier recommendation + apply-to-config; MVP preset gating. |
| `sentiment.rs` | `SentimentResult` + emotion/valence analysis via the local model. |
| `parse.rs` / `paths.rs` | Output parsing helpers / on-disk model artifact paths. |
| `local/` | Local runtime manager (was `local_ai/`). |
| `local/core.rs` | `LocalAiService` singleton (`global`/`try_global`), `model_artifact_path`. |
| `local/ops.rs` | Local RPC entrypoints (`local_ai_status/prompt/summarize/vision_prompt/embed/should_react`, `ReactionDecision`); re-exported as `local::rpc`. |
| `local/schemas.rs` | `local_ai.*` controller schemas + handlers. |
| `local/ollama.rs`, `local/lm_studio.rs` | Provider-specific runtime drivers; base-url resolution. |
| `local/install*.rs`, `local/voice_install_common.rs` | Whisper/Piper install + shared download logic. |
| `local/model_requirements.rs` | `MIN_CONTEXT_TOKENS`, `evaluate_context`, `ContextEligibility`. |
| `local/service/` | `LocalAiService` impl split: `bootstrap`, `ollama_admin`, `public_infer`, `speech`, `vision_embed`, `whisper_engine`, `assets`, `spawn_marker`. |
| `provider/` | Unified provider abstraction (was `providers/`). |
| `provider/traits.rs` | `Provider` trait + `ChatMessage`/`ChatRequest`/`ChatResponse`/`ToolCall`/`UsageInfo`/`ProviderDelta` etc. |
| `provider/factory.rs` | `create_chat_provider`, `provider_for_role`, provider-string grammar, local/cloud construction; `BYOK_INCOMPLETE_SENTINEL`. |
| `provider/router.rs` | `RouterProvider` hint-based multi-model routing. |
| `provider/reliable.rs` | Retry/backoff wrapper. |
| `provider/compatible*.rs` | OpenAI-compatible provider (request dump/parse/stream/types). |
| `provider/openhuman_backend.rs` | Managed OpenHuman backend provider (session JWT). |
| `provider/claude_agent_sdk/` | Claude Agent SDK subprocess provider (`protocol.rs`, `subprocess.rs`). |
| `provider/config_rejection.rs`, `provider/billing_error.rs` | Error classifiers (unknown-model / config rejection / budget exhausted). |
| `provider/temperature.rs`, `provider/thread_context.rs` | Per-workload temperature override; thread context plumbing. |
| `provider/ops.rs` | `list_configured_models`, SessionExpired publishing on auth failure. |
| `provider/schemas.rs` | Provider-layer schemas. |
| `voice/` | Inference implementations imported by `crate::openhuman::voice`. |
| `voice/cloud_transcribe.rs`, `voice/local_transcribe.rs`, `voice/local_speech.rs` | STT (cloud + local) and local TTS. |
| `voice/streaming.rs`, `voice/postprocess.rs`, `voice/hallucination.rs` | Streaming transcription, post-processing, hallucination filtering. |
| `openai_oauth/` | ChatGPT/Codex OAuth: `config.rs` (Codex OAuth config), `flow.rs` (start/complete/status/disconnect), `store.rs` (token persistence). |
| `http/` | OpenAI-compatible endpoint: `server.rs` (`router()`), `types.rs`; `EXTERNAL_OPENAI_COMPAT_PROVIDER` bearer id. |
| File / dir | Role |
| --------------------------------------------------------------------------------- | -------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| `mod.rs` | Domain root; module decls + re-exports; wires `inference.*` controller schemas/controllers. |
| `ops.rs` | Canonical handler file — `inference_*` business logic returning `RpcOutcome<T>`; delegates to `local`, `provider`, `sentiment`, `device`, `presets`, `openai_oauth`. Includes Sentry-noise suppression for expected provider/user-config failures. |
| `schemas.rs` | `inference.*` controller schemas + `handle_*` fns + param DTOs. |
| `types.rs` | Serde DTOs: `LocalAiStatus`, `LocalAiAssetsStatus`, `LocalAiDownloadsProgress`, `LocalAiEmbeddingResult`, `LocalAiSpeechResult`, `LocalAiTtsResult`, etc. |
| `device.rs` | `DeviceProfile` hardware detection (RAM/CPU/GPU/OS), cached. |
| `model_ids.rs` | Effective chat/vision/embedding/STT/TTS/quantization model id resolution from config. |
| `model_context.rs` | Known model context-window sizes (`context_window_for_model`) for pre-dispatch budgeting. |
| `presets.rs` | `ModelPreset`, `ModelTier`, `VisionMode`; tier recommendation + apply-to-config; MVP preset gating. |
| `sentiment.rs` | `SentimentResult` + emotion/valence analysis via the local model. |
| `parse.rs` / `paths.rs` | Output parsing helpers / on-disk model artifact paths. |
| `local/` | Local runtime manager (was `local_ai/`). |
| `local/core.rs` | `LocalAiService` singleton (`global`/`try_global`), `model_artifact_path`. |
| `local/ops.rs` | Local RPC entrypoints (`local_ai_status/prompt/summarize/vision_prompt/embed/should_react`, `ReactionDecision`); re-exported as `local::rpc`. |
| `local/schemas.rs` | `local_ai.*` controller schemas + handlers. |
| `local/ollama.rs`, `local/lm_studio.rs` | Provider-specific runtime drivers; base-url resolution. |
| `local/install*.rs`, `local/voice_install_common.rs` | Whisper/Piper install + shared download logic. |
| `local/model_requirements.rs` | `MIN_CONTEXT_TOKENS`, `evaluate_context`, `ContextEligibility`. |
| `local/service/` | `LocalAiService` impl split: `bootstrap`, `ollama_admin`, `public_infer`, `speech`, `vision_embed`, `whisper_engine`, `assets`, `spawn_marker`. |
| `provider/` | Unified provider abstraction (was `providers/`). |
| `provider/traits.rs` | `Provider` trait + `ChatMessage`/`ChatRequest`/`ChatResponse`/`ToolCall`/`UsageInfo`/`ProviderDelta` etc. |
| `provider/factory.rs` | `create_chat_provider`, `provider_for_role`, provider-string grammar, local/cloud construction; `BYOK_INCOMPLETE_SENTINEL`. |
| `provider/router.rs` | `RouterProvider` hint-based multi-model routing. |
| `provider/reliable.rs` | Retry/backoff wrapper. |
| `provider/compatible*.rs` | OpenAI-compatible provider (request dump/parse/stream/types). |
| `provider/openhuman_backend.rs` | Managed OpenHuman backend provider (session JWT). |
| `provider/claude_agent_sdk/` | Claude Agent SDK subprocess provider (`protocol.rs`, `subprocess.rs`). |
| `provider/config_rejection.rs`, `provider/billing_error.rs` | Error classifiers (unknown-model / config rejection / budget exhausted). |
| `provider/temperature.rs`, `provider/thread_context.rs` | Per-workload temperature override; thread context plumbing. |
| `provider/ops.rs` | `list_configured_models`, SessionExpired publishing on auth failure. |
| `provider/schemas.rs` | Provider-layer schemas. |
| `voice/` | Inference implementations imported by `crate::openhuman::voice`. |
| `voice/cloud_transcribe.rs`, `voice/local_transcribe.rs`, `voice/local_speech.rs` | STT (cloud + local) and local TTS. |
| `voice/streaming.rs`, `voice/postprocess.rs`, `voice/hallucination.rs` | Streaming transcription, post-processing, hallucination filtering. |
| `openai_oauth/` | ChatGPT/Codex OAuth: `config.rs` (Codex OAuth config), `flow.rs` (start/complete/status/disconnect), `store.rs` (token persistence). |
| `http/` | OpenAI-compatible endpoint: `server.rs` (`router()`), `types.rs`; `EXTERNAL_OPENAI_COMPAT_PROVIDER` bearer id. |
## Public surface
src/openhuman/threads/turn_state/mirror_tests.rs
+1
View File
@@ -225,6 +225,7 @@ fn subagent_lifecycle_records_and_clears_active() {
dedicated_thread: false,
prompt_chars: 42,
worker_thread_id: None,
display_name: None,
});
let s = m.snapshot();
assert_eq!(s.active_subagent.as_deref(), Some("researcher"));
tests/config_auth_app_state_connectivity_e2e.rs
+20
View File
@@ -651,6 +651,26 @@ fn config_schema_helpers_cover_provider_voice_agent_and_channel_defaults() {
migrate_legacy_fields(&mut custom_legacy);
assert_eq!(custom_legacy.label, "Custom");
assert!(custom_legacy.endpoint.is_empty());
let mut sumopod_legacy = CloudProviderCreds {
id: "provider-sumopod".to_string(),
legacy_type: Some("sumopod".to_string()),
..CloudProviderCreds::default()
};
migrate_legacy_fields(&mut sumopod_legacy);
assert_eq!(sumopod_legacy.slug, "sumopod");
assert_eq!(sumopod_legacy.label, "SumoPod");
assert_eq!(sumopod_legacy.endpoint, "https://ai.sumopod.com/v1");
assert_eq!(sumopod_legacy.auth_style, AuthStyle::Bearer);
let mut minimax_legacy = CloudProviderCreds {
id: "provider-minimax".to_string(),
legacy_type: Some("minimax".to_string()),
..CloudProviderCreds::default()
};
migrate_legacy_fields(&mut minimax_legacy);
assert_eq!(minimax_legacy.slug, "minimax");
assert_eq!(minimax_legacy.label, "MiniMax");
assert_eq!(minimax_legacy.endpoint, "https://api.minimax.io/anthropic");
assert_eq!(minimax_legacy.auth_style, AuthStyle::Anthropic);
assert_eq!(AuthStyle::OpenhumanJwt.as_str(), "openhuman_jwt");
assert_eq!(AuthStyle::Anthropic.as_str(), "anthropic");
assert_eq!(AuthStyle::None.as_str(), "none");
tests/memory_core_threads_raw_coverage_e2e.rs
+1
View File
@@ -554,6 +554,7 @@ async fn thread_ops_welcome_migration_and_turn_state_cover_error_and_cleanup_pat
dedicated_thread: true,
prompt_chars: 42,
worker_thread_id: None,
display_name: None,
}));
assert!(mirror.observe(&AgentProgress::SubagentCompleted {
agent_id: "researcher".into(),
tests/memory_threads_raw_coverage_e2e.rs
+1
View File
@@ -3140,6 +3140,7 @@ fn turn_state_mirror_persists_progress_edges_from_public_events() {
dedicated_thread: true,
prompt_chars: 99,
worker_thread_id: None,
display_name: None,
}));
assert!(!mirror.observe(&AgentProgress::SubagentIterationStarted {
agent_id: "researcher".into(),