mirrors/hermes-agent
1
0
Fork 0
mirror of https://github.com/NousResearch/hermes-agent.git synced 2026-05-05 02:31:47 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 6

feat: devex help, add Makefile, ruff, pre-commit, and modernize CI

Browse source
This commit is contained in:
Brooklyn Nicholson 2026-03-09 20:36:51 -05:00
parent 172a38c344
commit f4d7e6a29e
111 changed files with 11655 additions and 10200 deletions
Download patch file Download diff file Expand all files Collapse all files

26
agent/model_metadata.py
View file

@ -9,7 +9,7 @@ import os
import re
import time
from pathlib import Path
from typing import Any, Dict, List, Optional
from typing import Any
import requests
import yaml
@ -18,7 +18,7 @@ from hermes_constants import OPENROUTER_MODELS_URL
logger = logging.getLogger(__name__)
_model_metadata_cache: Dict[str, Dict[str, Any]] = {}
_model_metadata_cache: dict[str, dict[str, Any]] = {}
_model_metadata_cache_time: float = 0
_MODEL_CACHE_TTL = 3600
@ -63,7 +63,7 @@ DEFAULT_CONTEXT_LENGTHS = {
}
def fetch_model_metadata(force_refresh: bool = False) -> Dict[str, Dict[str, Any]]:
def fetch_model_metadata(force_refresh: bool = False) -> dict[str, dict[str, Any]]:
"""Fetch model metadata from OpenRouter (cached for 1 hour)."""
global _model_metadata_cache, _model_metadata_cache_time
@ -104,7 +104,7 @@ def _get_context_cache_path() -> Path:
return hermes_home / "context_length_cache.yaml"
def _load_context_cache() -> Dict[str, int]:
def _load_context_cache() -> dict[str, int]:
"""Load the model+provider → context_length cache from disk."""
path = _get_context_cache_path()
if not path.exists():
@ -139,14 +139,14 @@ def save_context_length(model: str, base_url: str, length: int) -> None:
logger.debug("Failed to save context length cache: %s", e)
def get_cached_context_length(model: str, base_url: str) -> Optional[int]:
def get_cached_context_length(model: str, base_url: str) -> int | None:
"""Look up a previously discovered context length for model+provider."""
key = f"{model}@{base_url}"
cache = _load_context_cache()
return cache.get(key)
def get_next_probe_tier(current_length: int) -> Optional[int]:
def get_next_probe_tier(current_length: int) -> int | None:
"""Return the next lower probe tier, or None if already at minimum."""
for tier in CONTEXT_PROBE_TIERS:
if tier < current_length:
@ -154,7 +154,7 @@ def get_next_probe_tier(current_length: int) -> Optional[int]:
return None
def parse_context_limit_from_error(error_msg: str) -> Optional[int]:
def parse_context_limit_from_error(error_msg: str) -> int | None:
"""Try to extract the actual context limit from an API error message.
Many providers include the limit in their error text, e.g.:
@ -166,11 +166,11 @@ def parse_context_limit_from_error(error_msg: str) -> Optional[int]:
error_lower = error_msg.lower()
# Pattern: look for numbers near context-related keywords
patterns = [
r'(?:max(?:imum)?|limit)\s*(?:context\s*)?(?:length|size|window)?\s*(?:is|of|:)?\s*(\d{4,})',
r'context\s*(?:length|size|window)\s*(?:is|of|:)?\s*(\d{4,})',
r'(\d{4,})\s*(?:token)?\s*(?:context|limit)',
r'>\s*(\d{4,})\s*(?:max|limit|token)', # "250000 tokens > 200000 maximum"
r'(\d{4,})\s*(?:max(?:imum)?)\b', # "200000 maximum"
r"(?:max(?:imum)?|limit)\s*(?:context\s*)?(?:length|size|window)?\s*(?:is|of|:)?\s*(\d{4,})",
r"context\s*(?:length|size|window)\s*(?:is|of|:)?\s*(\d{4,})",
r"(\d{4,})\s*(?:token)?\s*(?:context|limit)",
r">\s*(\d{4,})\s*(?:max|limit|token)", # "250000 tokens > 200000 maximum"
r"(\d{4,})\s*(?:max(?:imum)?)\b", # "200000 maximum"
]
for pattern in patterns:
match = re.search(pattern, error_lower)
@ -218,7 +218,7 @@ def estimate_tokens_rough(text: str) -> int:
return len(text) // 4
def estimate_messages_tokens_rough(messages: List[Dict[str, Any]]) -> int:
def estimate_messages_tokens_rough(messages: list[dict[str, Any]]) -> int:
"""Rough token estimate for a message list (pre-flight only)."""
total_chars = sum(len(str(msg)) for msg in messages)
return total_chars // 4