Spaces:
Running
Running
File size: 10,966 Bytes
bbe01fe 3d134a6 bbe01fe 3d134a6 bbe01fe 3d134a6 bbe01fe 3d134a6 bbe01fe 3d134a6 bbe01fe 3d134a6 bbe01fe 3d134a6 bbe01fe 3d134a6 bbe01fe 3d134a6 bbe01fe 3d134a6 bbe01fe 3d134a6 bbe01fe | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 | import json
import time
from typing import AsyncIterator, Literal, Optional, Protocol
import httpx
from groq import AsyncGroq
from tenacity import retry, stop_after_attempt, wait_fixed, retry_if_exception_type
from app.core.config import Settings
from app.core.exceptions import GenerationError
class TpmBucket:
"""
Sliding 60-second token-consumption tracker shared across all Groq calls.
Issue 7: When the bucket exceeds 12,000 estimated tokens in the current
minute window, complete_with_complexity() downgrades 70B calls to 8B
automatically. This leaves 2,400 TPM headroom and prevents hard failures
(HTTP 429) from degrading the service under load.
Token estimates are rough (prompt_chars / 4) but accurate enough for this
protective purpose — the goal is load shedding, not exact accounting.
"""
_WINDOW_SECONDS: int = 60
_DOWNGRADE_THRESHOLD: int = 12_000
def __init__(self) -> None:
self._count: int = 0
self._window_start: float = time.monotonic()
def add(self, estimated_tokens: int) -> None:
now = time.monotonic()
if now - self._window_start >= self._WINDOW_SECONDS:
self._count = 0
self._window_start = now
self._count += estimated_tokens
@property
def should_downgrade(self) -> bool:
now = time.monotonic()
if now - self._window_start >= self._WINDOW_SECONDS:
return False
return self._count > self._DOWNGRADE_THRESHOLD
class LLMClient(Protocol):
async def complete(self, prompt: str, system: str, stream: bool) -> AsyncIterator[str]:
...
async def classify_complexity(self, query: str) -> Literal["simple", "complex"]:
...
async def complete_with_complexity(self, prompt: str, system: str, stream: bool, complexity: str) -> AsyncIterator[str]:
...
class GroqClient:
def __init__(self, api_key: str, model_default: str, model_large: str, tpm_bucket: Optional[TpmBucket] = None):
if not api_key or api_key == "gsk_placeholder":
# We might be initialized in a test context without a real key
self.client = None
else:
self.client = AsyncGroq(api_key=api_key)
self.model_default = model_default
self.model_large = model_large
# Shared TPM bucket — injected at startup, None in test contexts.
self._tpm_bucket = tpm_bucket
@retry(stop=stop_after_attempt(2), wait=wait_fixed(1.0), retry=retry_if_exception_type((httpx.RequestError, httpx.TimeoutException)))
async def classify_complexity(self, query: str) -> Literal["simple", "complex"]:
if not self.client:
raise GenerationError("GroqClient not configured with an API Key.")
system = "You are a classifier. Read the user query. Output ONLY the word 'simple' or 'complex'. Do not explain."
try:
response = await self.client.chat.completions.create(
messages=[
{"role": "system", "content": system},
{"role": "user", "content": query}
],
model=self.model_default,
temperature=0.0,
max_tokens=10,
timeout=3.0,
)
result = response.choices[0].message.content.strip().lower()
if "complex" in result:
return "complex"
return "simple"
except Exception as e:
# Fallback to complex just to be safe if classification fails on parsing
return "complex"
@retry(stop=stop_after_attempt(2), wait=wait_fixed(1.0), retry=retry_if_exception_type((httpx.RequestError, httpx.TimeoutException)))
async def complete(self, prompt: str, system: str, stream: bool) -> AsyncIterator[str]:
if not self.client:
raise GenerationError("GroqClient not configured with an API Key.")
model = self.model_default
try:
stream_response = await self.client.chat.completions.create(
messages=[
{"role": "system", "content": system},
{"role": "user", "content": prompt}
],
model=model,
stream=True
)
async for chunk in stream_response:
content = chunk.choices[0].delta.content
if content:
yield content
except Exception as e:
raise GenerationError("Groq completion failed", context={"error": str(e)}) from e
async def complete_with_complexity(self, prompt: str, system: str, stream: bool, complexity: str) -> AsyncIterator[str]:
# Helper to allow pipeline nodes to pass the pre-classified complexity.
# Issue 7: if the shared TPM bucket is above 12,000 tokens in the current
# minute window, downgrade 70B to 8B to prevent hard rate-limit failures.
if not self.client:
raise GenerationError("GroqClient not configured with an API Key.")
if complexity == "complex" and self._tpm_bucket is not None and self._tpm_bucket.should_downgrade:
model = self.model_default
else:
model = self.model_large if complexity == "complex" else self.model_default
# Estimate input tokens before the call so the bucket reflects the full
# cost even when the response is long. 4 chars ≈ 1 token (rough heuristic).
if self._tpm_bucket is not None:
self._tpm_bucket.add((len(prompt) + len(system)) // 4)
try:
stream_response = await self.client.chat.completions.create(
messages=[
{"role": "system", "content": system},
{"role": "user", "content": prompt}
],
model=model,
stream=stream # Instruct strictly said stream=True yields token chunks.
)
if stream:
async for chunk in stream_response:
content = chunk.choices[0].delta.content
if content:
# Accumulate estimated response tokens in the bucket.
if self._tpm_bucket is not None:
self._tpm_bucket.add(len(content) // 4 or 1)
yield content
else:
full = stream_response.choices[0].message.content
if self._tpm_bucket is not None and full:
self._tpm_bucket.add(len(full) // 4)
yield full
except Exception as e:
raise GenerationError("Groq completion failed", context={"error": str(e)}) from e
class OllamaClient:
def __init__(self, base_url: str, model: str):
self.base_url = base_url.rstrip("/")
self.model = model
@retry(stop=stop_after_attempt(2), wait=wait_fixed(1.0), retry=retry_if_exception_type((httpx.RequestError, httpx.TimeoutException)))
async def classify_complexity(self, query: str) -> Literal["simple", "complex"]:
system = "You are a classifier. Read the user query. Output ONLY the word 'simple' or 'complex'. Do not explain."
try:
async with httpx.AsyncClient() as client:
response = await client.post(
f"{self.base_url}/api/chat",
json={
"model": self.model,
"messages": [
{"role": "system", "content": system},
{"role": "user", "content": query}
],
"stream": False,
"options": {
"temperature": 0.0,
"num_predict": 10
}
},
timeout=3.0
)
response.raise_for_status()
data = response.json()
result = data.get("message", {}).get("content", "").strip().lower()
if "complex" in result:
return "complex"
return "simple"
except Exception:
return "complex"
@retry(stop=stop_after_attempt(2), wait=wait_fixed(1.0), retry=retry_if_exception_type((httpx.RequestError, httpx.TimeoutException)))
async def complete(self, prompt: str, system: str, stream: bool) -> AsyncIterator[str]:
async with httpx.AsyncClient() as client:
try:
async with client.stream(
"POST",
f"{self.base_url}/api/chat",
json={
"model": self.model,
"messages": [
{"role": "system", "content": system},
{"role": "user", "content": prompt}
],
"stream": True # Force true per instruction
}
) as response:
response.raise_for_status()
async for line in response.aiter_lines():
if line:
try:
data = json.loads(line)
if "message" in data and "content" in data["message"]:
yield data["message"]["content"]
except json.JSONDecodeError:
pass
except Exception as e:
raise GenerationError("Ollama completion failed", context={"error": str(e)}) from e
async def complete_with_complexity(self, prompt: str, system: str, stream: bool, complexity: str) -> AsyncIterator[str]:
# Ollama just uses one model in this implementation
async for token in self.complete(prompt, system, stream):
yield token
def get_llm_client(settings: Settings, tpm_bucket: Optional[TpmBucket] = None) -> LLMClient:
if settings.LLM_PROVIDER == "ollama":
if not settings.OLLAMA_BASE_URL or not settings.OLLAMA_MODEL:
raise ValueError("OLLAMA_BASE_URL and OLLAMA_MODEL must be explicitly set when LLM_PROVIDER is 'ollama'")
return OllamaClient(
base_url=settings.OLLAMA_BASE_URL,
model=settings.OLLAMA_MODEL
)
else:
# Defaults to Groq
return GroqClient(
api_key=settings.GROQ_API_KEY or "",
model_default=settings.GROQ_MODEL_DEFAULT,
model_large=settings.GROQ_MODEL_LARGE,
tpm_bucket=tpm_bucket,
)
|