Hitonabi/mission-control-v2
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Refactor: Pricing/Draft-Pfad als Single Source of Truth (Phase 1)

Browse Source 
- Neuer services/pricing.py: PRICING-Dict + compute_savings() aus dem
  system-Router extrahiert; Router ist jetzt dünn (nur role_map + Aufruf).
- /system/token-stats liefert zusätzlich das pricing-Dict → Frontend zeigt
  die Tarife daraus an statt sie im Text zu hartkodieren.
- SPEC_DRAFT_MODEL_PATH in config.py (MC_SPEC_DRAFT_MODEL); llamaswap.py und
  migrate_config.py referenzieren die Konstante statt des doppelten Literals.
- Ersparnis-Berechnung verhaltensneutral verifiziert (35,09 $ / 32,28 €).

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
This commit is contained in:
Hitonabi
2026-06-26 14:28:26 +02:00
parent 35dcc69ba5
commit a7c3f8f516
8 changed files with 155 additions and 133 deletions
Download Patch File Download Diff File Expand all files Collapse all files
backend/config.py
+4
View File
@@ -30,6 +30,10 @@ CMD_TEMPLATE = os.environ.get("MC_CMD_TEMPLATE", _DEFAULT_CMD_TEMPLATE)
if "{model}" not in CMD_TEMPLATE:
CMD_TEMPLATE = _DEFAULT_CMD_TEMPLATE
DEFAULT_TTL = int(os.environ.get("MC_DEFAULT_TTL", "300"))
# Draft-Modell für Speculative Decoding (nur fast/coder, wenn vorhanden). Eine
# Quelle der Wahrheit für llamaswap.register_model + migrate_config.
SPEC_DRAFT_MODEL_PATH = os.environ.get(
"MC_SPEC_DRAFT_MODEL", f"{MODELS_DIR.as_posix()}/drafts/qwen2.5-1.5b-instruct-q4_k_m.gguf")
# Env für HuggingFace-Downloads: XET deaktivieren (Hänger bei ~6 MB, siehe v1-Gotcha).
HF_DOWNLOAD_ENV = {"HF_HUB_DISABLE_XET": "1"}
backend/migrate_config.py
+4 -4
View File
@@ -5,7 +5,7 @@ from pathlib import Path
sys.path.append(str(Path(__file__).resolve().parent))
from services.llamaswap import read_config, write_config
from config import CONFIG_PATH
from config import CONFIG_PATH, SPEC_DRAFT_MODEL_PATH
def migrate():
print(f"Reading config from {CONFIG_PATH}...")
@@ -15,9 +15,9 @@ def migrate():
cfg = read_config()
models = cfg.get("models", {})
draft_path = "/srv/models/drafts/qwen2.5-1.5b-instruct-q4_k_m.gguf"
draft_path = SPEC_DRAFT_MODEL_PATH
for name, spec in models.items():
if not isinstance(spec, dict):
continue
backend/routers/system.py
+11 -54
View File
@@ -5,6 +5,7 @@ Lokal (Windows) schlagen die Shell-Befehle harmlos fehl und werden als Fehler
zurückgegeben statt zu crashen.
"""
import logging
import os
import subprocess
@@ -15,8 +16,12 @@ from config import GATEWAY_URL, HERMES_API_URL, HERMES_WEBUI_URL, LLAMA_SWAP_URL
from services import backup as backup_svc
from services.agent import agent_status
from services.gateway import gateway_reachable
from services.llamaswap import engine_reachable
from services.llamaswap import engine_reachable, list_models
from services.pricing import compute_savings
from services.system import system_status
from services.token_stats import get_stats
log = logging.getLogger(__name__)
router = APIRouter(prefix="/api")
@@ -90,64 +95,16 @@ def self_update() -> dict:
return {"pull": pull, "reset": reset, "restart": restart_res}
from services.token_stats import get_stats
from services.llamaswap import list_models
@router.get("/system/token-stats")
def token_stats() -> dict:
stats = get_stats()
p = stats.get("prompt_tokens", 0)
c = stats.get("completion_tokens", 0)
total = p + c
# Map model IDs and aliases to their respective roles for pricing resolution
role_map = {}
"""Token-Verbrauch + Cloud-Ersparnis. Logik im pricing-Service (SSoT)."""
# Rolle je Modell/Alias (lowercase) für die Tarif-Auflösung auflösen.
role_map: dict[str, str | None] = {}
try:
for m in list_models():
role_map[m["name"].lower()] = m.get("role")
for alias in m.get("aliases", []):
role_map[alias.lower()] = m.get("role")
except Exception:
pass
# Dynamic pricing tiers based on model class in June 2026
PRICING = {
"heavy": (15.0, 75.0),
"coder": (3.0, 15.0),
"hermes": (1.0, 5.0),
"fast": (0.15, 0.60),
"scout": (0.15, 0.60),
"vision": (0.15, 0.60),
"reasoning": (0.15, 0.60),
}
modeled_p = 0
modeled_c = 0
saved_usd = 0.0
models_data = stats.get("models") or {}
for m_name, m_tokens in models_data.items():
mp = m_tokens.get("prompt", 0)
mc = m_tokens.get("completion", 0)
modeled_p += mp
modeled_c += mc
role = role_map.get(m_name, m_name)
rate_in, rate_out = PRICING.get(role, (0.15, 0.60))
saved_usd += (mp * rate_in + mc * rate_out) / 1_000_000.0
# Baseline/legacy tokens calculated at premium rates ($15.00 / $75.00)
# to preserve historical savings value prior to model-specific logging
baseline_p = max(0, p - modeled_p)
baseline_c = max(0, c - modeled_c)
saved_usd += (baseline_p * 15.0 + baseline_c * 75.0) / 1_000_000.0
saved_eur = saved_usd * 0.92 # 1 USD = 0.92 EUR
return {
"prompt_tokens": p,
"completion_tokens": c,
"total_tokens": total,
"saved_usd": round(saved_usd, 2),
"saved_eur": round(saved_eur, 2)
}
log.warning("token_stats: list_models fehlgeschlagen, Tarife per Name", exc_info=True)
return compute_savings(get_stats(), role_map)
backend/services/llamaswap.py
+3 -4
View File
@@ -12,7 +12,7 @@ import re
import httpx
from ruamel.yaml.scalarstring import LiteralScalarString
from config import CMD_TEMPLATE, CONFIG_PATH, DEFAULT_TTL, LLAMA_SWAP_URL
from config import CMD_TEMPLATE, CONFIG_PATH, DEFAULT_TTL, LLAMA_SWAP_URL, SPEC_DRAFT_MODEL_PATH
# Kanonische Rollen (vereinheitlicht ggü. v1: kein manager/reviewer mehr).
ROLE_IDS = {"vision", "coder", "reasoning", "agent", "scout"}
@@ -188,9 +188,8 @@ def register_model(model_path: str, role: str | None = None, ctx: int = 8192,
if role_lower in ("fast", "coder"):
if "--parallel" not in cmd:
cmd += " --parallel 2"
draft_path = "/srv/models/drafts/qwen2.5-1.5b-instruct-q4_k_m.gguf"
if os.path.exists(draft_path) and "--spec-draft-model" not in cmd:
cmd += f" --spec-draft-model {draft_path}"
if os.path.exists(SPEC_DRAFT_MODEL_PATH) and "--spec-draft-model" not in cmd:
cmd += f" --spec-draft-model {SPEC_DRAFT_MODEL_PATH}"
cfg.setdefault("models", {})[model_id] = {
"cmd": LiteralScalarString(cmd + "\n"),
backend/services/pricing.py
+58
View File
@@ -0,0 +1,58 @@
"""Kosten-/Ersparnis-Berechnung für die Token-Statistik (eine Quelle der Wahrheit).
Vergleicht die lokal verbrauchten Tokens gegen die Cloud-Listenpreise vergleichbarer
Modellklassen (Stand Juni 2026, USD pro 1M Tokens, in/out) und liefert die so
eingesparte Summe. Wird vom System-Router dünn aufgerufen.
"""
import os
# Cloud-Listenpreise je Rolle/Modellklasse: (input_usd_per_1M, output_usd_per_1M).
PRICING: dict[str, tuple[float, float]] = {
"heavy": (15.0, 75.0),
"coder": (3.0, 15.0),
"hermes": (1.0, 5.0),
"fast": (0.15, 0.60),
"scout": (0.15, 0.60),
"vision": (0.15, 0.60),
"reasoning": (0.15, 0.60),
}
# Tarif für nicht zuordenbare Tokens (Default-/Fallback-Klasse).
DEFAULT_RATE: tuple[float, float] = (0.15, 0.60)
# Baseline/Legacy-Tokens (vor modellspezifischem Logging) am Premium-Tarif bewerten,
# damit historische Ersparnis erhalten bleibt.
BASELINE_RATE: tuple[float, float] = PRICING["heavy"]
USD_TO_EUR = float(os.environ.get("MC_USD_TO_EUR", "0.92"))
def compute_savings(stats: dict, role_map: dict[str, str | None]) -> dict:
"""Aggregiert Tokens und berechnet die Cloud-Ersparnis.
role_map: Modell-/Alias-Name (lowercase) -> Rolle, zur Tarif-Auflösung.
"""
prompt = stats.get("prompt_tokens", 0)
completion = stats.get("completion_tokens", 0)
modeled_p = modeled_c = 0
saved_usd = 0.0
for m_name, m_tokens in (stats.get("models") or {}).items():
mp = m_tokens.get("prompt", 0)
mc = m_tokens.get("completion", 0)
modeled_p += mp
modeled_c += mc
role = role_map.get(m_name, m_name)
rate_in, rate_out = PRICING.get(role, DEFAULT_RATE)
saved_usd += (mp * rate_in + mc * rate_out) / 1_000_000.0
baseline_p = max(0, prompt - modeled_p)
baseline_c = max(0, completion - modeled_c)
saved_usd += (baseline_p * BASELINE_RATE[0] + baseline_c * BASELINE_RATE[1]) / 1_000_000.0
return {
"prompt_tokens": prompt,
"completion_tokens": completion,
"total_tokens": prompt + completion,
"saved_usd": round(saved_usd, 2),
"saved_eur": round(saved_usd * USD_TO_EUR, 2),
"pricing": {role: {"in": r[0], "out": r[1]} for role, r in PRICING.items()},
}