xsl
54dc741811
DeepSeek (OpenAI-compatible) as primary with a 智谱 GLM slot that is a real routing path but raises until the 算法备案 model list is confirmed. Adds scene routing, failover, a per-provider circuit breaker, structured output with retry-on-validation-failure, prompts loaded by id+version from prompts/, and cost attribution written to ob_ai_call for every attempt -- successes, retries and failures alike. Two controls exist because they are cheap now and expensive later: a cache-prefix validator that rejects timestamps, uuids and nicknames in the cached system prefix (a volatile prefix silently multiplies input cost), and a hard platform-wide daily budget that rejects non-critical scenes instead of merely alerting, since a retry storm outruns any human on-call.
65 lines
1.8 KiB
Python
Executable File
65 lines
1.8 KiB
Python
Executable File
"""LLM gateway — 多供应商路由、Prompt 版本、结构化输出、成本计量、预算保护.
|
|
|
|
架构文档 §4.3: V1 是进程内包，V2 拆为独立服务。接口按远程调用设计，拆分零成本。
|
|
"""
|
|
|
|
from packages.llm.budget import DailyBudgetGuard, InMemoryCostCounter, RedisCostCounter
|
|
from packages.llm.circuit import CircuitBreaker, CircuitBreakerRegistry, CircuitState
|
|
from packages.llm.cost import (
|
|
AICallRecord,
|
|
AICallRecorder,
|
|
CompositeAICallRecorder,
|
|
InMemoryAICallRecorder,
|
|
ModelRate,
|
|
NullAICallRecorder,
|
|
RateTable,
|
|
compute_cost_cents,
|
|
)
|
|
from packages.llm.errors import (
|
|
AllProvidersFailedError,
|
|
CachePrefixViolationError,
|
|
DailyBudgetExceededError,
|
|
LLMError,
|
|
ProviderError,
|
|
StructuredOutputError,
|
|
)
|
|
from packages.llm.gateway import LLMGateway
|
|
from packages.llm.prompts import PromptRegistry, PromptTemplate, validate_cache_prefix
|
|
from packages.llm.routing import SceneRouter, build_scene_configs
|
|
from packages.llm.types import Completion, Message, PromptRef, Scene, SceneConfig, Usage
|
|
|
|
__all__ = [
|
|
"AICallRecord",
|
|
"AICallRecorder",
|
|
"AllProvidersFailedError",
|
|
"CachePrefixViolationError",
|
|
"CircuitBreaker",
|
|
"CircuitBreakerRegistry",
|
|
"CircuitState",
|
|
"Completion",
|
|
"CompositeAICallRecorder",
|
|
"DailyBudgetExceededError",
|
|
"DailyBudgetGuard",
|
|
"InMemoryAICallRecorder",
|
|
"InMemoryCostCounter",
|
|
"LLMError",
|
|
"LLMGateway",
|
|
"Message",
|
|
"ModelRate",
|
|
"NullAICallRecorder",
|
|
"PromptRef",
|
|
"PromptRegistry",
|
|
"PromptTemplate",
|
|
"ProviderError",
|
|
"RateTable",
|
|
"RedisCostCounter",
|
|
"Scene",
|
|
"SceneConfig",
|
|
"SceneRouter",
|
|
"StructuredOutputError",
|
|
"Usage",
|
|
"build_scene_configs",
|
|
"compute_cost_cents",
|
|
"validate_cache_prefix",
|
|
]
|