Compare commits

..
Author SHA1 Message Date
Shroominic dd4ed7541f undo 43e9732 2025-12-22 11:31:19 +01:00
Shroominic cc42534a97 manual alias due to openrouter api bug 2025-12-22 11:30:51 +01:00
9qeklajc 43e97326e0 fix model naming issue in response 2025-12-12 20:14:43 +01:00
9qeklajc 06770a0702 Merge branch 'v0.2.1' into embeddings-integration 2025-12-12 19:44:45 +01:00
Shroominic 547365894d add simple test 2025-12-11 13:58:57 +08:00
Shroominic c0176a5274 Merge branch 'v0.2.1' into embeddings-integration 2025-12-11 13:06:20 +08:00
Shroominic 329d22363f Merge branch 'v0.2.1' into embeddings-integration 2025-12-11 12:59:41 +08:00
9qeklajc 9438bc957f clean up 2025-12-03 22:24:43 +01:00
9qeklajc 5d2219880d embedding integration 2025-12-03 22:23:03 +01:00
9qeklajc 195da0c9da embedding integration 2025-12-03 22:22:57 +01:00
12 changed files with 315 additions and 536 deletions
+1 -121
View File
@@ -5,7 +5,7 @@ from pathlib import Path
from fastapi import APIRouter, Depends, HTTPException, Query, Request
from fastapi.responses import HTMLResponse, RedirectResponse
from pydantic import BaseModel, Field
from pydantic import BaseModel
from sqlmodel import select
from ..payment.models import _row_to_model, list_models
@@ -3165,123 +3165,3 @@ async def get_log_dates_api(request: Request) -> dict[str, object]:
continue
return {"dates": dates}
class ModelMappingRequest(BaseModel):
from_model: str = Field(..., alias="from")
to: str
class ModelMappingUpdateRequest(BaseModel):
to: str
@admin_router.get("/api/model-mappings", dependencies=[Depends(require_admin_api)])
async def get_model_mappings(request: Request) -> dict[str, str]:
from ..proxy import _manual_model_mappings
return _manual_model_mappings
@admin_router.post("/api/model-mappings", dependencies=[Depends(require_admin_api)])
async def create_model_mapping(request: Request, mapping: ModelMappingRequest) -> dict[str, str]:
import json
import os
from ..proxy import _manual_model_mappings, load_manual_model_mappings
mappings_file = os.path.join(os.path.dirname(os.path.dirname(__file__)), "model_mappings.json")
try:
if os.path.exists(mappings_file):
with open(mappings_file, "r") as f:
data = json.load(f)
else:
data = {"manual_model_mappings": {"mappings": {}}}
data["manual_model_mappings"]["mappings"][mapping.from_model.lower()] = mapping.to.lower()
with open(mappings_file, "w") as f:
json.dump(data, f, indent=2)
load_manual_model_mappings()
return _manual_model_mappings
except Exception as e:
raise HTTPException(status_code=500, detail=f"Failed to create mapping: {str(e)}")
@admin_router.put("/api/model-mappings/{from_model}", dependencies=[Depends(require_admin_api)])
async def update_model_mapping(request: Request, from_model: str, mapping: ModelMappingUpdateRequest) -> dict[str, str]:
import json
import os
from ..proxy import _manual_model_mappings, load_manual_model_mappings
mappings_file = os.path.join(os.path.dirname(os.path.dirname(__file__)), "model_mappings.json")
try:
if os.path.exists(mappings_file):
with open(mappings_file, "r") as f:
data = json.load(f)
else:
data = {"manual_model_mappings": {"mappings": {}}}
if from_model.lower() not in data["manual_model_mappings"]["mappings"]:
raise HTTPException(status_code=404, detail="Mapping not found")
data["manual_model_mappings"]["mappings"][from_model.lower()] = mapping.to.lower()
with open(mappings_file, "w") as f:
json.dump(data, f, indent=2)
load_manual_model_mappings()
return _manual_model_mappings
except HTTPException:
raise
except Exception as e:
raise HTTPException(status_code=500, detail=f"Failed to update mapping: {str(e)}")
@admin_router.delete("/api/model-mappings/{from_model}", dependencies=[Depends(require_admin_api)])
async def delete_model_mapping(request: Request, from_model: str) -> dict[str, str]:
import json
import os
from ..proxy import _manual_model_mappings, load_manual_model_mappings
mappings_file = os.path.join(os.path.dirname(os.path.dirname(__file__)), "model_mappings.json")
try:
if os.path.exists(mappings_file):
with open(mappings_file, "r") as f:
data = json.load(f)
else:
data = {"manual_model_mappings": {"mappings": {}}}
if from_model.lower() not in data["manual_model_mappings"]["mappings"]:
raise HTTPException(status_code=404, detail="Mapping not found")
del data["manual_model_mappings"]["mappings"][from_model.lower()]
with open(mappings_file, "w") as f:
json.dump(data, f, indent=2)
load_manual_model_mappings()
return _manual_model_mappings
except HTTPException:
raise
except Exception as e:
raise HTTPException(status_code=500, detail=f"Failed to delete mapping: {str(e)}")
@admin_router.post("/api/model-mappings/reload", dependencies=[Depends(require_admin_api)])
async def reload_model_mappings(request: Request) -> dict[str, object]:
from ..proxy import _manual_model_mappings, load_manual_model_mappings
try:
load_manual_model_mappings()
return {"ok": True, "mappings": _manual_model_mappings}
except Exception as e:
raise HTTPException(status_code=500, detail=f"Failed to reload mappings: {str(e)}")
-7
View File
@@ -1,7 +0,0 @@
{
"manual_model_mappings": {
"mappings": {
"text-embedding-ada-002-v2": "text-embedding-ada-002"
}
}
}
+2 -62
View File
@@ -5,7 +5,6 @@ from pydantic.v1 import BaseModel
from ..core import get_logger
from ..core.db import AsyncSession
from ..core.settings import settings
from .price import sats_usd_price
logger = get_logger(__name__)
@@ -65,56 +64,6 @@ async def calculate_cost( # todo: can be sync
)
return cost_data
usage_data = response_data["usage"]
usd_cost = 0.0
# Prioritize cost_details.upstream_inference_cost
if "cost_details" in usage_data:
usd_cost = float(
usage_data["cost_details"].get("upstream_inference_cost", 0) or 0
)
# Fallback to cost field if upstream_inference_cost is 0
if usd_cost == 0 and "cost" in usage_data:
try:
usd_cost = float(usage_data.get("cost", 0) or 0)
except Exception:
pass
if usd_cost > 0:
try:
sats_per_usd = 1.0 / sats_usd_price()
cost_in_sats = usd_cost * sats_per_usd
cost_in_msats = math.ceil(cost_in_sats * 1000)
logger.info(
"Using cost from usage data/details",
extra={
"usd_cost": usd_cost,
"cost_in_sats": cost_in_sats,
"cost_in_msats": cost_in_msats,
"model": response_data.get("model", "unknown"),
},
)
return CostData(
base_msats=-1,
input_msats=-1, # Cost field doesn't break down by token type
output_msats=-1,
total_msats=cost_in_msats,
)
except Exception as e:
logger.warning(
"Error calculating cost from usage data",
extra={
"error": str(e),
"usd_cost": usd_cost,
"model": response_data.get("model", "unknown"),
},
)
# Fall through to token-based calculation
MSATS_PER_1K_INPUT_TOKENS: float = (
float(settings.fixed_per_1k_input_tokens) * 1000.0
)
@@ -180,19 +129,10 @@ async def calculate_cost( # todo: can be sync
)
return cost_data
input_tokens = usage_data.get("prompt_tokens", 0)
output_tokens = usage_data.get("completion_tokens", 0)
# added for response api
input_tokens = (
input_tokens if input_tokens != 0 else usage_data.get("input_tokens", 0)
)
output_tokens = (
output_tokens if output_tokens != 0 else usage_data.get("output_tokens", 0)
)
input_tokens = response_data.get("usage", {}).get("prompt_tokens", 0)
output_tokens = response_data.get("usage", {}).get("completion_tokens", 0)
input_msats = round(input_tokens / 1000 * MSATS_PER_1K_INPUT_TOKENS, 3)
output_msats = round(output_tokens / 1000 * MSATS_PER_1K_OUTPUT_TOKENS, 3)
token_based_cost = math.ceil(input_msats + output_msats)
+4 -43
View File
@@ -1,5 +1,4 @@
import json
import os
from typing import Any
from fastapi import APIRouter, Depends, HTTPException, Request
@@ -34,23 +33,6 @@ _upstreams: list[BaseUpstreamProvider] = []
_model_instances: dict[str, Model] = {} # All aliases -> Model
_provider_map: dict[str, BaseUpstreamProvider] = {} # All aliases -> Provider
_unique_models: dict[str, Model] = {} # Unique model.id -> Model (no duplicates)
_manual_model_mappings: dict[str, str] = {} # Manual model_id mappings loaded from JSON
def load_manual_model_mappings() -> None:
"""Load manual model mappings from JSON file."""
global _manual_model_mappings
try:
mappings_file = os.path.join(os.path.dirname(__file__), "model_mappings.json")
if os.path.exists(mappings_file):
with open(mappings_file, "r") as f:
data = json.load(f)
_manual_model_mappings = data.get("manual_model_mappings", {}).get("mappings", {})
else:
_manual_model_mappings = {}
except Exception as e:
logger.error(f"Failed to load manual model mappings: {e}")
_manual_model_mappings = {}
async def initialize_upstreams() -> None:
@@ -58,7 +40,6 @@ async def initialize_upstreams() -> None:
global _upstreams
_upstreams = await init_upstreams()
logger.info(f"Initialized {len(_upstreams)} upstream providers")
load_manual_model_mappings()
await refresh_model_maps()
@@ -70,7 +51,6 @@ async def reinitialize_upstreams() -> None:
"Re-initialized upstream providers from admin action",
extra={"provider_count": len(_upstreams)},
)
load_manual_model_mappings()
await refresh_model_maps()
@@ -84,32 +64,13 @@ def get_upstreams() -> list[BaseUpstreamProvider]:
def get_model_instance(model_id: str) -> Model | None:
"""Get Model instance by ID from global cache, with manual mapping fallback."""
model = _model_instances.get(model_id)
if model is not None:
return model
mapped_model_id = _manual_model_mappings.get(model_id.lower())
if mapped_model_id:
return _model_instances.get(mapped_model_id.lower())
return None
"""Get Model instance by ID from global cache."""
return _model_instances.get(model_id.lower())
def get_provider_for_model(model_id: str) -> BaseUpstreamProvider | None:
"""Get UpstreamProvider for model ID from global cache, with manual mapping fallback."""
# First try direct lookup
provider = _provider_map.get(model_id)
if provider is not None:
return provider
# Try manual mapping as fallback
mapped_model_id = _manual_model_mappings.get(model_id)
if mapped_model_id:
logger.debug(f"Using manual mapping for provider: {model_id} -> {mapped_model_id}")
return _provider_map.get(mapped_model_id)
return None
"""Get UpstreamProvider for model ID from global cache."""
return _provider_map.get(model_id)
def get_unique_models() -> list[Model]:
-1
View File
@@ -734,7 +734,6 @@ class BaseUpstreamProvider:
await client.aclose()
return mapped_error
# Handle endpoints that require cost calculation and payment adjustment
if path.endswith("chat/completions") or path.endswith("embeddings"):
if path.endswith("chat/completions"):
client_wants_streaming = False
+7 -1
View File
@@ -50,7 +50,13 @@ class OpenRouterUpstreamProvider(BaseUpstreamProvider):
async def fetch_models(self) -> list[Model]:
"""Fetch all OpenRouter models."""
models_data = await async_fetch_openrouter_models()
return [Model(**model) for model in models_data] # type: ignore
models = [Model(**model) for model in models_data] # type: ignore
# manual alias for openai/text-embedding-ada-002 due to openrouter api bug
for model in models:
if model.id == "openai/text-embedding-ada-002":
model.alias_ids = ["text-embedding-ada-002-v2"]
break
return models
async def get_balance(self) -> float | None:
"""Get the current account balance from OpenRouter.
+97
View File
@@ -0,0 +1,97 @@
import json
from typing import Any
from unittest.mock import AsyncMock, MagicMock, patch
import pytest
from httpx import AsyncClient
@pytest.mark.integration
@pytest.mark.asyncio
async def test_proxy_embeddings_endpoint(authenticated_client: AsyncClient) -> None:
"""Test the embeddings endpoint proxy functionality"""
test_payload = {
"model": "text-embedding-ada-002",
"input": "The quick brown fox",
}
mock_response_data = {
"object": "list",
"data": [
{"object": "embedding", "embedding": [0.0023, -0.0012, 0.0045], "index": 0}
],
"model": "text-embedding-ada-002",
"usage": {"prompt_tokens": 5, "total_tokens": 5},
}
with patch("httpx.AsyncClient.send") as mock_send:
# Create a proper async generator for iter_bytes
async def mock_iter_bytes(*args: Any, **kwargs: Any) -> Any:
yield json.dumps(mock_response_data).encode()
mock_response = AsyncMock()
mock_response.status_code = 200
mock_response.headers = {"content-type": "application/json"}
mock_response.text = json.dumps(mock_response_data)
# Use MagicMock for synchronous .json() method
mock_response.json = MagicMock(return_value=mock_response_data)
mock_response.iter_bytes = mock_iter_bytes
mock_response.aiter_bytes = mock_iter_bytes
mock_send.return_value = mock_response
# Make POST request to embeddings endpoint
response = await authenticated_client.post("/v1/embeddings", json=test_payload)
assert response.status_code == 200
response_data = response.json()
assert response_data["object"] == "list"
assert len(response_data["data"]) == 1
assert response_data["data"][0]["object"] == "embedding"
# Verify request was forwarded
mock_send.assert_called_once()
forwarded_request = mock_send.call_args[0][0]
# Verify the path ends with embeddings
# Note: forwarded path might be full URL
assert str(forwarded_request.url).endswith("embeddings")
@pytest.mark.integration
@pytest.mark.asyncio
async def test_model_case_insensitivity(authenticated_client: AsyncClient) -> None:
"""Test that model lookups are case insensitive"""
# We'll use a mixed-case model ID that should match the lowercase one in the system
# We assume 'gpt-3.5-turbo' is available in the mock env/database
test_payload = {
"model": "GPT-3.5-TURBO",
"messages": [{"role": "user", "content": "Hello"}],
}
with patch("httpx.AsyncClient.send") as mock_send:
mock_response_data = {
"id": "chatcmpl-123",
"object": "chat.completion",
"choices": [{"message": {"content": "Hi"}}],
"usage": {"total_tokens": 10},
}
async def mock_iter_bytes(*args: Any, **kwargs: Any) -> Any:
yield json.dumps(mock_response_data).encode()
mock_response = AsyncMock()
mock_response.status_code = 200
mock_response.headers = {"content-type": "application/json"}
mock_response.text = json.dumps(mock_response_data)
mock_response.json = MagicMock(return_value=mock_response_data)
mock_response.iter_bytes = mock_iter_bytes
mock_response.aiter_bytes = mock_iter_bytes
mock_send.return_value = mock_response
response = await authenticated_client.post(
"/v1/chat/completions", json=test_payload
)
assert response.status_code == 200
@@ -3,6 +3,7 @@ Integration tests for wallet authentication system including API key generation
Tests POST /v1/wallet/topup endpoint and authorization header validation.
"""
import hashlib
from datetime import datetime, timedelta
from typing import Any
@@ -14,6 +15,7 @@ from routstr.core.db import ApiKey
from .utils import (
CashuTokenGenerator,
ConcurrencyTester,
ResponseValidator,
)
@@ -387,6 +389,69 @@ async def test_api_key_with_expiry_time(
# The expiry time and refund address functionality is tested elsewhere
@pytest.mark.integration
@pytest.mark.asyncio
async def test_concurrent_token_submissions(
integration_client: AsyncClient, testmint_wallet: Any, integration_session: Any
) -> None:
"""Test concurrent submissions of different tokens"""
# Generate multiple unique tokens with known amounts
num_tokens = 10
tokens = []
expected_balances = {}
for i in range(num_tokens):
amount = 100 + i * 10
token = await testmint_wallet.mint_tokens(amount)
tokens.append(token)
# Store expected balance by token hash
hashed_key = hashlib.sha256(token.encode()).hexdigest()
expected_balances[hashed_key] = amount * 1000 # msats
# Create concurrent requests
requests = [
{
"method": "GET",
"url": "/v1/wallet/info",
"headers": {"Authorization": f"Bearer {token}"},
}
for token in tokens
]
# Execute concurrently
tester = ConcurrencyTester()
responses = await tester.run_concurrent_requests(
integration_client, requests, max_concurrent=5
)
# All should succeed
assert len(responses) == num_tokens
api_keys = set()
for response in responses:
assert response.status_code == 200
data = response.json()
api_key = data["api_key"]
api_keys.add(api_key)
# Verify balance matches the expected amount
hashed_key = api_key[3:] # Remove "sk-" prefix
assert data["balance"] == expected_balances[hashed_key]
# Should have created unique API keys
assert len(api_keys) == num_tokens
# Verify all keys exist in database
for api_key in api_keys:
hashed_key = api_key[3:] # Remove "sk-" prefix
result = await integration_session.execute(
select(ApiKey).where(ApiKey.hashed_key == hashed_key) # type: ignore[arg-type]
)
db_key = result.scalar_one()
assert db_key.balance == expected_balances[hashed_key]
@pytest.mark.integration
@pytest.mark.asyncio
async def test_authorization_with_cashu_token_directly(
@@ -439,6 +504,48 @@ async def test_x_cashu_header_support(
assert response.status_code == 200
@pytest.mark.integration
@pytest.mark.asyncio
@pytest.mark.slow
async def test_api_key_consistency_under_load(
integration_client: AsyncClient, testmint_wallet: Any, integration_session: Any
) -> None:
"""Test API key generation consistency under concurrent load"""
# Generate a single token
token = await testmint_wallet.mint_tokens(1000)
# First request to create the API key
integration_client.headers["Authorization"] = f"Bearer {token}"
initial_response = await integration_client.get("/v1/wallet/info")
assert initial_response.status_code == 200
expected_api_key = initial_response.json()["api_key"]
expected_balance = initial_response.json()["balance"]
# Try to use the same token concurrently multiple times
# All should return the same API key since it's already created
requests = [
{
"method": "GET",
"url": "/v1/wallet/info",
"headers": {"Authorization": f"Bearer {token}"},
}
for _ in range(20) # 20 concurrent attempts
]
tester = ConcurrencyTester()
responses = await tester.run_concurrent_requests(
integration_client, requests, max_concurrent=10
)
# All should succeed and return the same API key
for response in responses:
assert response.status_code == 200
data = response.json()
assert data["api_key"] == expected_api_key
assert data["balance"] == expected_balance
@pytest.mark.integration
@pytest.mark.asyncio
async def test_database_timestamp_accuracy(
+40 -1
View File
@@ -13,7 +13,7 @@ from sqlmodel import select, update
from routstr.core.db import ApiKey
from .utils import ResponseValidator
from .utils import ConcurrencyTester, ResponseValidator
@pytest.mark.integration
@@ -204,6 +204,45 @@ async def test_expired_api_key_behavior(
assert db_key.refund_address == "test@lightning.address"
@pytest.mark.integration
@pytest.mark.asyncio
async def test_concurrent_access_same_api_key(
integration_client: AsyncClient, authenticated_client: AsyncClient
) -> None:
"""Test concurrent access with the same API key"""
# Get the API key from authenticated client
response = await authenticated_client.get("/v1/wallet/")
api_key = response.json()["api_key"]
initial_balance = response.json()["balance"]
# Create multiple concurrent requests
requests = []
for i in range(20):
# Alternate between both endpoints
endpoint = "/v1/wallet/" if i % 2 == 0 else "/v1/wallet/info"
requests.append(
{
"method": "GET",
"url": endpoint,
"headers": {"Authorization": f"Bearer {api_key}"},
}
)
# Execute concurrently
tester = ConcurrencyTester()
responses = await tester.run_concurrent_requests(
integration_client, requests, max_concurrent=10
)
# All should succeed with consistent data
for response in responses:
assert response.status_code == 200
data = response.json()
assert data["api_key"] == api_key
assert data["balance"] == initial_balance
@pytest.mark.integration
@pytest.mark.asyncio
async def test_wallet_info_data_consistency(
+55
View File
@@ -15,6 +15,7 @@ from routstr.core.db import ApiKey
from .utils import (
CashuTokenGenerator,
ConcurrencyTester,
ResponseValidator,
)
@@ -283,6 +284,60 @@ async def test_transaction_history_tracking( # type: ignore[no-untyped-def]
assert response.status_code == 400
@pytest.mark.integration
@pytest.mark.asyncio
async def test_concurrent_topups_same_api_key( # type: ignore[no-untyped-def]
integration_client: AsyncClient,
authenticated_client: AsyncClient,
testmint_wallet: Any,
) -> None:
"""Test concurrent top-ups to the same API key"""
# Get API key
response = await authenticated_client.get("/v1/wallet/")
api_key = response.json()["api_key"]
initial_balance = response.json()["balance"]
# Generate multiple unique tokens
num_tokens = 10
tokens = []
total_amount = 0
for i in range(num_tokens):
amount = 100 + i * 10 # Different amounts
token = await testmint_wallet.mint_tokens(amount)
tokens.append(token)
total_amount += amount
# Create concurrent top-up requests
requests = [
{
"method": "POST",
"url": "/v1/wallet/topup",
"params": {"cashu_token": token},
"headers": {"Authorization": f"Bearer {api_key}"},
}
for token in tokens
]
# Execute concurrently
tester = ConcurrencyTester()
responses = await tester.run_concurrent_requests(
integration_client, requests, max_concurrent=5
)
# All should succeed
for response in responses:
assert response.status_code == 200
assert "msats" in response.json()
# Verify final balance is correct
final_response = await authenticated_client.get("/v1/wallet/")
final_balance = final_response.json()["balance"]
expected_balance = initial_balance + (total_amount * 1000)
assert final_balance == expected_balance
@pytest.mark.integration
@pytest.mark.asyncio
async def test_topup_during_active_proxy_request( # type: ignore[no-untyped-def]
+2 -222
View File
@@ -10,26 +10,16 @@ import { SiteHeader } from '@/components/site-header';
import { Tabs, TabsContent, TabsList, TabsTrigger } from '@/components/ui/tabs';
import { useQuery } from '@tanstack/react-query';
import { AdminService } from '@/lib/api/services/admin';
import { ModelMappingService } from '@/lib/api/services/modelMappings';
import { Skeleton } from '@/components/ui/skeleton';
import { AlertCircle, Users, Globe } from 'lucide-react';
import { Alert, AlertDescription } from '@/components/ui/alert';
import { Badge } from '@/components/ui/badge';
import React, { useMemo, useState } from 'react';
import { useMemo, useState } from 'react';
import type { Model } from '@/lib/api/schemas/models';
import { groupAndSortModelsByProvider } from '@/lib/utils/modelSort';
import { Button } from '@/components/ui/button';
import { Input } from '@/components/ui/input';
import { Card, CardContent, CardHeader, CardTitle } from '@/components/ui/card';
import { Trash2, Plus, Edit2, Save, X } from 'lucide-react';
export default function ModelsPage() {
const [filteredModels, setFilteredModels] = useState<Model[]>([]);
const [modelMappings, setModelMappings] = useState<Record<string, string>>(
{}
);
const [editingMapping, setEditingMapping] = useState<string | null>(null);
const [newMapping, setNewMapping] = useState({ from: '', to: '' });
const {
data: modelsData,
@@ -41,23 +31,6 @@ export default function ModelsPage() {
refetchOnWindowFocus: false,
});
const {
data: mappingsData,
isLoading: isLoadingMappings,
error: mappingsError,
refetch: refetchMappings,
} = useQuery({
queryKey: ['model-mappings'],
queryFn: () => ModelMappingService.getModelMappings(),
refetchOnWindowFocus: false,
});
React.useEffect(() => {
if (mappingsData) {
setModelMappings(mappingsData);
}
}, [mappingsData]);
const { models = [], groups = [] } = modelsData || {};
const groupedModels = useMemo(() => {
@@ -94,40 +67,6 @@ export default function ModelsPage() {
});
}, [groupedModels, groupDataMap, groups]);
const handleAddMapping = async () => {
if (!newMapping.from || !newMapping.to) return;
try {
await ModelMappingService.createModelMapping({
from: newMapping.from,
to: newMapping.to,
});
setNewMapping({ from: '', to: '' });
refetchMappings();
} catch (error) {
console.error('Failed to add mapping:', error);
}
};
const handleDeleteMapping = async (from: string) => {
try {
await ModelMappingService.deleteModelMapping(from);
refetchMappings();
} catch (error) {
console.error('Failed to delete mapping:', error);
}
};
const handleUpdateMapping = async (from: string, to: string) => {
try {
await ModelMappingService.updateModelMapping(from, { to });
setEditingMapping(null);
refetchMappings();
} catch (error) {
console.error('Failed to update mapping:', error);
}
};
return (
<SidebarProvider>
<AppSidebar variant='inset' />
@@ -142,9 +81,8 @@ export default function ModelsPage() {
</div>
<Tabs defaultValue='manage' className='w-full'>
<TabsList className='grid w-full grid-cols-4'>
<TabsList className='grid w-full grid-cols-3'>
<TabsTrigger value='manage'>Manage Models</TabsTrigger>
<TabsTrigger value='mappings'>Model Mappings</TabsTrigger>
{/*<TabsTrigger value='test-basic'>Basic Testing</TabsTrigger>
<TabsTrigger value='test-api'>API Endpoints</TabsTrigger> */}
</TabsList>
@@ -329,164 +267,6 @@ export default function ModelsPage() {
)}
</TabsContent>
<TabsContent value='mappings' className='space-y-4'>
<div className='text-muted-foreground text-sm'>
Manage model ID mappings to redirect requests from one model
to another. This is useful for maintaining compatibility with
legacy model names or creating aliases.
</div>
{isLoadingMappings ? (
<div className='space-y-4'>
<Skeleton className='h-[200px] w-full' />
</div>
) : mappingsError ? (
<Alert variant='destructive'>
<AlertCircle className='h-4 w-4' />
<AlertDescription>
Failed to load model mappings. Please try refreshing the
page.
</AlertDescription>
</Alert>
) : (
<div className='space-y-6'>
<Card>
<CardHeader>
<CardTitle className='flex items-center gap-2'>
<Plus className='h-5 w-5' />
Add New Model Mapping
</CardTitle>
</CardHeader>
<CardContent>
<div className='grid grid-cols-1 gap-4 md:grid-cols-3'>
<Input
placeholder='From model ID'
value={newMapping.from}
onChange={(e) =>
setNewMapping({
...newMapping,
from: e.target.value,
})
}
/>
<Input
placeholder='To model ID'
value={newMapping.to}
onChange={(e) =>
setNewMapping({
...newMapping,
to: e.target.value,
})
}
/>
<Button
onClick={handleAddMapping}
disabled={!newMapping.from || !newMapping.to}
className='w-full'
>
<Plus className='mr-2 h-4 w-4' />
Add Mapping
</Button>
</div>
</CardContent>
</Card>
<Card>
<CardHeader>
<CardTitle>Current Model Mappings</CardTitle>
</CardHeader>
<CardContent>
{Object.keys(modelMappings).length === 0 ? (
<div className='text-muted-foreground py-8 text-center'>
No model mappings configured
</div>
) : (
<div className='space-y-3'>
{Object.entries(modelMappings).map(([from, to]) => (
<div
key={from}
className='flex items-center justify-between gap-4 rounded-lg border p-4'
>
<div className='grid flex-1 grid-cols-1 gap-4 md:grid-cols-2'>
<div>
<label className='text-muted-foreground text-sm font-medium'>
From
</label>
<div className='font-mono text-sm'>
{from}
</div>
</div>
<div>
<label className='text-muted-foreground text-sm font-medium'>
To
</label>
{editingMapping === from ? (
<div className='flex items-center gap-2'>
<Input
defaultValue={to}
id={`edit-${from}`}
className='text-sm'
/>
<Button
size='sm'
onClick={() => {
const input =
document.getElementById(
`edit-${from}`
) as HTMLInputElement;
handleUpdateMapping(
from,
input.value
);
}}
>
<Save className='h-4 w-4' />
</Button>
<Button
size='sm'
variant='outline'
onClick={() =>
setEditingMapping(null)
}
>
<X className='h-4 w-4' />
</Button>
</div>
) : (
<div className='font-mono text-sm'>
{to}
</div>
)}
</div>
</div>
{editingMapping !== from && (
<div className='flex items-center gap-2'>
<Button
size='sm'
variant='outline'
onClick={() => setEditingMapping(from)}
>
<Edit2 className='h-4 w-4' />
</Button>
<Button
size='sm'
variant='destructive'
onClick={() => handleDeleteMapping(from)}
>
<Trash2 className='h-4 w-4' />
</Button>
</div>
)}
</div>
))}
</div>
)}
</CardContent>
</Card>
</div>
)}
</TabsContent>
<TabsContent value='test-basic' className='space-y-4'>
<div className='text-muted-foreground text-sm'>
Test model credentials and connectivity with basic chat
-78
View File
@@ -1,78 +0,0 @@
import { apiClient } from '../client';
import { z } from 'zod';
export const ModelMappingSchema = z.object({
from: z.string(),
to: z.string(),
});
export const CreateModelMappingSchema = z.object({
from: z.string(),
to: z.string(),
});
export const UpdateModelMappingSchema = z.object({
to: z.string(),
});
export const ModelMappingsResponseSchema = z.record(z.string());
export const ReloadMappingsResponseSchema = z.object({
ok: z.boolean(),
mappings: z.record(z.string()),
});
export type ModelMapping = z.infer<typeof ModelMappingSchema>;
export type CreateModelMapping = z.infer<typeof CreateModelMappingSchema>;
export type UpdateModelMapping = z.infer<typeof UpdateModelMappingSchema>;
export type ModelMappingsResponse = z.infer<typeof ModelMappingsResponseSchema>;
export type ReloadMappingsResponse = z.infer<
typeof ReloadMappingsResponseSchema
>;
export class ModelMappingService {
static async getModelMappings(): Promise<ModelMappingsResponse> {
return await apiClient.get<ModelMappingsResponse>(
'/admin/api/model-mappings'
);
}
static async createModelMapping(
data: CreateModelMapping
): Promise<ModelMappingsResponse> {
return await apiClient.post<ModelMappingsResponse>(
'/admin/api/model-mappings',
{
from: data.from,
to: data.to,
}
);
}
static async updateModelMapping(
fromModel: string,
data: UpdateModelMapping
): Promise<ModelMappingsResponse> {
return await apiClient.put<ModelMappingsResponse>(
`/admin/api/model-mappings/${encodeURIComponent(fromModel)}`,
{
to: data.to,
}
);
}
static async deleteModelMapping(
fromModel: string
): Promise<ModelMappingsResponse> {
return await apiClient.delete<ModelMappingsResponse>(
`/admin/api/model-mappings/${encodeURIComponent(fromModel)}`
);
}
static async reloadModelMappings(): Promise<ReloadMappingsResponse> {
return await apiClient.post<ReloadMappingsResponse>(
'/admin/api/model-mappings/reload',
{}
);
}
}