diff --git a/TEST_SUITE_ANALYSIS.md b/TEST_SUITE_ANALYSIS.md
new file mode 100644
index 00000000..8e4b6196
--- /dev/null
+++ b/TEST_SUITE_ANALYSIS.md
@@ -0,0 +1,524 @@
+# Comprehensive Test Suite Analysis Report
+
+**Date:** 2024  
+**Target Audience:** Senior pytest engineer with full codebase understanding  
+**Purpose:** Evaluate test suite effectiveness, identify outdated tests, and highlight missing coverage
+
+---
+
+## Executive Summary
+
+This report analyzes the routstr test suite (27 test files, ~210 test functions) against the current codebase implementation. The analysis reveals:
+
+- **Outdated Tests:** 15+ tests that no longer match current implementation
+- **Low-Value Tests:** 8+ tests that don't provide meaningful coverage
+- **Missing Coverage:** Critical gaps in reserved balance, cost calculation, model mapping, and admin functionality
+- **Test Quality:** Generally good structure but needs modernization to match recent architectural changes
+
+---
+
+## 1. Outdated Tests (No Longer Valid)
+
+### 1.1 GET Request Billing Tests
+
+**Files:** `test_proxy_get_endpoints.py`
+
+**Issue:** Multiple tests assume GET requests are billed, but the current implementation (`proxy.py:191-202`) explicitly allows unauthenticated GET requests without billing.
+
+**Affected Tests:**
+- `test_proxy_get_billing_verification` (lines 236-277)
+- `test_proxy_get_billing_calculations_match_pricing` (lines 326-365)
+- `test_proxy_get_database_state_verification` (lines 369-416)
+- `test_proxy_get_insufficient_balance` (lines 281-322) - GET requests succeed even with low balance
+
+**Recommendation:** Remove billing assertions from GET tests or refactor to test actual behavior (GET requests are free).
+
+**Code Reference:** `proxy.py:191-202` - GET requests bypass authentication and billing
+
+---
+
+### 1.2 Reserved Balance Tests
+
+**Files:** `test_reserved_balance_negative.py`, `test_database_consistency.py`
+
+**Issue:** Tests reference `reserved_balance` field which exists in the codebase (`auth.py:304-344`) but tests don't properly exercise the reservation/release flow.
+
+**Affected Tests:**
+- `test_reserved_balance_never_negative` - Tests exist but don't cover concurrent reservation scenarios
+- `test_balance_update_atomicity` - Doesn't test reserved balance atomicity
+
+**Recommendation:** Update tests to properly test the reserved balance lifecycle:
+1. Reservation during `pay_for_request`
+2. Release/charge during `adjust_payment_for_tokens`
+3. Revert during `revert_pay_for_request`
+
+**Code Reference:** `auth.py:304-344` (reservation), `auth.py:387-420` (revert), `auth.py:423-662` (adjustment)
+
+---
+
+### 1.3 Model Pricing Storage Tests
+
+**Files:** `test_fee_consistency.py`
+
+**Issue:** Tests verify pricing is stored without fees, but the current implementation applies provider fees when reading from database (`proxy.py:93-99`, `payment/models.py:_row_to_model`).
+
+**Affected Tests:**
+- `test_pricing_stored_without_fees` - Correctly tests storage, but misses fee application on read
+- Missing: Tests for provider fee application during model retrieval
+
+**Recommendation:** Add tests for fee application during model retrieval, not just storage.
+
+**Code Reference:** `payment/models.py:_row_to_model` applies `provider_fee` multiplier
+
+---
+
+### 1.4 Background Task Tests
+
+**Files:** `test_background_tasks.py`
+
+**Issue:** Several tests are skipped or test functionality that doesn't exist.
+
+**Affected Tests:**
+- `test_executes_at_configured_intervals` (line 443) - Skipped, tests timing
+- `test_calculates_payouts_accurately` (line 448) - Tests `periodic_payout` which just logs a warning
+- `test_refund_check_disabled` (line 424) - Commented out, tests disabled functionality
+
+**Recommendation:** 
+- Remove tests for unimplemented features (`periodic_payout`)
+- Implement proper tests for `refresh_model_maps_periodically` (exists in `proxy.py:116-131`)
+
+**Code Reference:** `proxy.py:116-131` - `refresh_model_maps_periodically` exists but untested
+
+---
+
+### 1.5 Provider Management Tests
+
+**Files:** `test_provider_management.py`
+
+**Issue:** Tests mock Nostr relay queries but don't test actual NIP-91 parsing logic.
+
+**Affected Tests:**
+- All tests mock `query_nostr_relay_for_providers` - Should test actual `nip91.py` parsing
+
+**Recommendation:** Add unit tests for `nip91.py` parsing logic, integration tests can mock relay but should test parser.
+
+**Code Reference:** `nip91.py` - NIP-91 event parsing logic exists but untested
+
+---
+
+## 2. Low-Value Tests (Not Testing Useful Behavior)
+
+### 2.1 Over-Mocked Integration Tests
+
+**Files:** `test_proxy_post_endpoints.py`, `test_proxy_get_endpoints.py`
+
+**Issue:** Tests mock `httpx.AsyncClient.send` so extensively that they don't test actual proxy logic.
+
+**Affected Tests:**
+- `test_proxy_post_json_payload_forwarding` - Mocks entire upstream response
+- `test_proxy_get_response_streaming` - Mocks streaming, doesn't test actual stream handling
+
+**Recommendation:** Use real upstream mocks (like the Docker testmint) or test at a higher level. Current tests verify mocks work, not the proxy.
+
+---
+
+### 2.2 Trivial Assertion Tests
+
+**Files:** `test_wallet_authentication.py`
+
+**Issue:** Some tests just verify basic structure without testing edge cases.
+
+**Affected Tests:**
+- `test_api_key_generation_valid_token` - Just checks response structure, doesn't test concurrent creation
+- `test_database_timestamp_accuracy` - Tests that key exists, not timestamp accuracy (timestamp field doesn't exist)
+
+**Recommendation:** Remove trivial tests or enhance them to test actual behavior (concurrency, edge cases).
+
+---
+
+### 2.3 Skipped Tests That Should Be Removed
+
+**Files:** `test_background_tasks.py`, `test_database_consistency.py`
+
+**Issue:** Tests marked with `@pytest.mark.skip` with reasons like "skipping for CI reliability" - these should either be fixed or removed.
+
+**Affected Tests:**
+- `test_executes_at_configured_intervals` (line 443)
+- `test_balance_never_negative` (line 382) - Marked skip with "Balance never negative is not implemented"
+
+**Recommendation:** Remove skipped tests if functionality doesn't exist, or fix them if it does.
+
+---
+
+### 2.4 Tests That Don't Match Implementation
+
+**Files:** `test_proxy_post_insufficient_balance` (line 637)
+
+**Issue:** Test is skipped with comment "depends on model pricing configuration" but the actual implementation (`auth.py:304-325`) does check balance and raises 402.
+
+**Recommendation:** Fix and enable this test - it should work with proper model setup.
+
+---
+
+## 3. Missing Critical Test Coverage
+
+### 3.1 Reserved Balance System
+
+**Priority: CRITICAL**
+
+**Missing Coverage:**
+- Concurrent reservation race conditions
+- Reservation release during successful requests
+- Reservation revert during failed requests
+- Reserved balance going negative (should be prevented)
+- Multiple concurrent requests with same API key
+
+**Code to Test:**
+- `auth.py:pay_for_request` (lines 289-384) - Reservation logic
+- `auth.py:revert_pay_for_request` (lines 387-420) - Revert logic
+- `auth.py:adjust_payment_for_tokens` (lines 423-662) - Adjustment logic
+
+**Recommended Tests:**
+```python
+async def test_concurrent_reservation_prevents_double_spend()
+async def test_reserved_balance_released_on_success()
+async def test_reserved_balance_reverted_on_upstream_failure()
+async def test_reserved_balance_never_negative()
+async def test_multiple_concurrent_requests_same_key()
+```
+
+---
+
+### 3.2 Cost Calculation Logic
+
+**Priority: CRITICAL**
+
+**Missing Coverage:**
+- `calculate_cost` function (`payment/cost_caculation.py`) - No tests exist
+- Token-based cost calculation
+- Max cost fallback logic
+- Cost adjustment based on actual usage
+- Image token calculation integration with cost
+
+**Code to Test:**
+- `payment/cost_caculation.py:calculate_cost` - Core cost calculation
+- `auth.py:adjust_payment_for_tokens` - Payment adjustment based on cost
+
+**Recommended Tests:**
+```python
+async def test_calculate_cost_with_token_usage()
+async def test_calculate_cost_max_cost_fallback()
+async def test_cost_adjustment_refunds_excess()
+async def test_cost_adjustment_charges_additional()
+async def test_cost_calculation_with_image_tokens()
+```
+
+---
+
+### 3.3 Model Mapping Algorithm
+
+**Priority: HIGH**
+
+**Missing Coverage:**
+- `create_model_mappings` function (`algorithm.py:168-300`) - Only unit tests for helper functions
+- Model override application
+- Disabled model filtering
+- Provider fee application in mappings
+- OpenRouter penalty logic in practice
+
+**Code to Test:**
+- `algorithm.py:create_model_mappings` - Main mapping function
+- `proxy.py:refresh_model_maps` - Refresh logic
+
+**Recommended Tests:**
+```python
+async def test_create_model_mappings_with_overrides()
+async def test_create_model_mappings_filters_disabled()
+async def test_model_mapping_applies_provider_fees()
+async def test_openrouter_penalty_in_mapping()
+async def test_refresh_model_maps_updates_cache()
+```
+
+---
+
+### 3.4 Upstream Provider System
+
+**Priority: HIGH**
+
+**Missing Coverage:**
+- Provider initialization (`upstream/helpers.py:init_upstreams`)
+- Provider-specific request forwarding
+- Provider error handling
+- X-Cashu authentication flow per provider
+- Provider header preparation
+
+**Code to Test:**
+- `upstream/helpers.py:init_upstreams`
+- `upstream/base.py:forward_request`
+- `upstream/base.py:handle_x_cashu`
+- Provider-specific implementations (OpenAI, Anthropic, etc.)
+
+**Recommended Tests:**
+```python
+async def test_init_upstreams_loads_all_providers()
+async def test_provider_forward_request_handles_errors()
+async def test_x_cashu_authentication_per_provider()
+async def test_provider_header_preparation()
+async def test_provider_specific_error_mapping()
+```
+
+---
+
+### 3.5 Admin Endpoints
+
+**Priority: MEDIUM**
+
+**Missing Coverage:**
+- Admin authentication (`core/admin.py`)
+- Model override CRUD operations
+- Provider management endpoints
+- Settings management
+- Admin password management
+
+**Code to Test:**
+- `core/admin.py` - All admin endpoints (lines 1-2800+)
+
+**Recommended Tests:**
+```python
+async def test_admin_authentication_required()
+async def test_create_model_override()
+async def test_update_model_override()
+async def test_delete_model_override()
+async def test_provider_crud_operations()
+async def test_settings_update()
+```
+
+---
+
+### 3.6 Discovery and NIP-91
+
+**Priority: MEDIUM**
+
+**Missing Coverage:**
+- NIP-91 event parsing (`nip91.py`)
+- Provider discovery from Nostr relay
+- Provider health checking
+- Provider caching logic
+
+**Code to Test:**
+- `nip91.py` - NIP-91 parsing
+- `discovery.py` - Discovery logic
+
+**Recommended Tests:**
+```python
+def test_parse_nip91_event_valid()
+def test_parse_nip91_event_malformed()
+async def test_query_nostr_relay_for_providers()
+async def test_fetch_provider_health()
+async def test_provider_cache_expiration()
+```
+
+---
+
+### 3.7 Error Handling Edge Cases
+
+**Priority: MEDIUM**
+
+**Missing Coverage:**
+- Cost calculation errors (`CostDataError`)
+- Upstream timeout handling
+- Partial stream failures
+- Database connection failures during payment
+- Mint communication failures during topup
+
+**Code to Test:**
+- `auth.py:adjust_payment_for_tokens` - CostDataError handling (line 634)
+- `upstream/base.py` - Timeout handling
+- `wallet.py:credit_balance` - Mint failure handling
+
+**Recommended Tests:**
+```python
+async def test_cost_calculation_error_handling()
+async def test_upstream_timeout_returns_error()
+async def test_partial_stream_failure_refunds()
+async def test_database_failure_during_payment()
+async def test_mint_unavailable_during_topup()
+```
+
+---
+
+### 3.8 Background Task Interactions
+
+**Priority: LOW**
+
+**Missing Coverage:**
+- Model map refresh during active requests
+- Pricing update during request processing
+- Task cancellation and cleanup
+- Task error recovery
+
+**Code to Test:**
+- `proxy.py:refresh_model_maps_periodically`
+- `payment/models.py:update_sats_pricing`
+
+**Recommended Tests:**
+```python
+async def test_model_refresh_during_active_requests()
+async def test_pricing_update_doesnt_break_requests()
+async def test_background_task_cancellation()
+async def test_background_task_error_recovery()
+```
+
+---
+
+## 4. Test Structure Issues
+
+### 4.1 Test Organization
+
+**Issues:**
+- Unit tests mix with integration concerns (e.g., `test_payment_helpers.py` uses mocks but tests integration logic)
+- Integration tests are well-organized but some test files are too large (900+ lines)
+
+**Recommendation:**
+- Split large test files (`test_proxy_post_endpoints.py`, `test_background_tasks.py`)
+- Separate unit tests from integration tests more clearly
+- Create dedicated test files for each major module
+
+---
+
+### 4.2 Fixture Usage
+
+**Issues:**
+- `conftest.py` has good fixtures but some tests create their own mocks instead of using fixtures
+- `testmint_wallet` fixture is good but some tests bypass it
+
+**Recommendation:**
+- Standardize on fixtures for common test scenarios
+- Create more reusable fixtures for common mock scenarios
+
+---
+
+### 4.3 Test Data Management
+
+**Issues:**
+- Tests create test data inline instead of using factories
+- Database state management is inconsistent
+
+**Recommendation:**
+- Create test data factories for common entities (ApiKey, Model, Provider)
+- Standardize database cleanup between tests
+
+---
+
+## 5. Specific Test Recommendations
+
+### 5.1 Tests to Remove
+
+1. **`test_proxy_get_billing_verification`** - GET requests aren't billed
+2. **`test_proxy_get_billing_calculations_match_pricing`** - GET requests aren't billed
+3. **`test_proxy_get_insufficient_balance`** - GET requests succeed without balance
+4. **`test_calculates_payouts_accurately`** - Feature not implemented
+5. **`test_balance_never_negative`** - Marked as not implemented, should be removed or implemented
+6. **`test_database_timestamp_accuracy`** - Tests non-existent timestamp field
+
+### 5.2 Tests to Fix
+
+1. **`test_proxy_post_insufficient_balance`** - Should work, needs proper model setup
+2. **`test_reserved_balance_never_negative`** - Should test actual reservation logic
+3. **All provider management tests** - Should test actual NIP-91 parsing, not just mocked responses
+
+### 5.3 Tests to Add (Priority Order)
+
+**Critical:**
+1. Reserved balance concurrent operations
+2. Cost calculation with various scenarios
+3. Payment adjustment logic (refund/excess charge)
+4. Model mapping with overrides and disabled models
+
+**High:**
+5. Upstream provider initialization and error handling
+6. X-Cashu authentication flow
+7. Admin endpoint authentication and CRUD
+
+**Medium:**
+8. NIP-91 parsing unit tests
+9. Discovery service integration
+10. Background task error recovery
+
+---
+
+## 6. Code Coverage Gaps
+
+Based on code analysis, these modules have low/no test coverage:
+
+1. **`payment/cost_caculation.py`** - 0% coverage (no tests exist)
+2. **`algorithm.py:create_model_mappings`** - Only helper functions tested
+3. **`upstream/helpers.py:init_upstreams`** - Not tested
+4. **`core/admin.py`** - Minimal/no tests
+5. **`nip91.py`** - Not tested
+6. **`discovery.py`** - Only mocked integration tests
+7. **`payment/lnurl.py`** - Not tested
+
+---
+
+## 7. Recommendations Summary
+
+### Immediate Actions (High Priority)
+
+1. **Remove outdated GET billing tests** - They test non-existent behavior
+2. **Add reserved balance tests** - Critical for payment correctness
+3. **Add cost calculation tests** - Core business logic untested
+4. **Fix/enable skipped tests** - Either fix or remove them
+
+### Short-term Actions (Medium Priority)
+
+5. **Add model mapping integration tests** - Test actual mapping behavior
+6. **Add upstream provider tests** - Test provider initialization and error handling
+7. **Add admin endpoint tests** - Critical for admin functionality
+8. **Refactor over-mocked tests** - Use real mocks or test at higher level
+
+### Long-term Actions (Low Priority)
+
+9. **Reorganize test structure** - Split large files, improve organization
+10. **Add NIP-91 unit tests** - Test parsing logic independently
+11. **Improve test fixtures** - Standardize and reuse fixtures
+12. **Add background task tests** - Test task interactions and error recovery
+
+---
+
+## 8. Test Quality Metrics
+
+**Current State:**
+- Total test functions: ~210
+- Unit tests: 7 files, ~30 tests
+- Integration tests: 20 files, ~180 tests
+- Skipped tests: ~5
+- Tests with issues: ~25
+
+**Coverage Estimate:**
+- Core payment logic: ~60% (missing cost calculation)
+- Authentication: ~70% (missing edge cases)
+- Proxy logic: ~50% (over-mocked)
+- Admin: ~10% (minimal tests)
+- Discovery: ~30% (only mocked)
+
+**Target State:**
+- Increase coverage to 80%+ for critical paths
+- Reduce skipped tests to 0
+- Fix all outdated tests
+- Add missing critical tests
+
+---
+
+## Conclusion
+
+The test suite has a solid foundation but needs significant updates to match the current codebase architecture. The most critical gaps are in reserved balance handling, cost calculation, and model mapping - all core business logic. Many tests are outdated due to architectural changes (GET request billing removal, reserved balance introduction).
+
+**Priority Focus Areas:**
+1. Reserved balance system (CRITICAL)
+2. Cost calculation logic (CRITICAL)
+3. Remove/fix outdated tests (HIGH)
+4. Model mapping tests (HIGH)
+5. Admin endpoint tests (MEDIUM)
+
+This report should serve as a roadmap for test suite modernization and expansion.