feat(learn): metered embeddings and the cache (decision-engine 6.2)

plane/models/openrouter-embedding.yaml: openai/text-embedding-3-small via
OpenRouter (embeddings only), input_per_m 0.02, cache/output 0, egress
openrouter.ai; `openrouter` joins the provider extensions.

plane/budget.py MeteredEmbeddingClient: prompt_tokens priced from the
Model manifest, usd/calls ceilings stop the next call, a priced call that
reports no tokens raises UnmeteredUsage (a silent zero is an instrument
failure). OpenRouterEmbeddingClient keeps one usage row per call
(parse_embedding_payload); its api key is out of repr.

learn/embeddings.py + `ousast learn memory embed`: excerpts embedded in
batches into embeddings/<model-slug>/<excerpt_sha>.json (float32 base64),
cached by content hash so a rerun makes no call; the declared dimension is
checked; --dry-run estimates tokens and cost without a call; a paid run
needs --budget-usd and exits 3 (unfinished) at its ceiling. Not run against
the provider: the store embedding waits for its budget go-ahead.

Co-Authored-By: Claude Opus 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_01EmM1j46qgKMh3zqD1QMwEo
This commit is contained in:
Marius Ciepluch
2026-10-01 09:56:52 +02:00
co-authored by Claude Opus 5.5
parent 7b81c9ac4e
commit 98fa4ec8cd
11 changed files with 498 additions and 11 deletions
+13 -3
View File
@@ -8,7 +8,7 @@ import time
import urllib.error
import urllib.request
from collections.abc import Callable, Mapping, Sequence
from dataclasses import dataclass
from dataclasses import dataclass, field
from typing import TypeVar, cast
@@ -135,10 +135,11 @@ class OpenRouterChatClient:
@dataclass(frozen=True)
class OpenRouterEmbeddingClient:
api_key: str
api_key: str = field(repr=False)
base_url: str = "https://openrouter.ai/api/v1"
max_attempts: int = 3
retry_base_delay: float = 0.5
usage: list[dict[str, object]] = field(default_factory=list, compare=False, repr=False) # one per call, for the meter
@classmethod
def from_env(cls) -> OpenRouterEmbeddingClient:
@@ -167,7 +168,16 @@ class OpenRouterEmbeddingClient:
response_payload = call_with_retry(_do, attempts=self.max_attempts, base_delay=self.retry_base_delay)
except (urllib.error.URLError, TimeoutError, http.client.HTTPException, json.JSONDecodeError) as exc:
raise OpenRouterError(f"OpenRouter embedding request failed: {type(exc).__name__}: {exc}") from exc
return parse_embedding_response(response_payload)
vectors, usage = parse_embedding_payload(response_payload)
self.usage.append(usage)
return vectors
def parse_embedding_payload(payload: object) -> tuple[list[list[float]], dict[str, object]]:
"""The vectors and the provider's ``usage`` (``{}`` when absent: the meter treats that as unmetered, not free)."""
vectors = parse_embedding_response(payload)
usage = payload.get("usage") if isinstance(payload, dict) else None
return vectors, dict(usage) if isinstance(usage, dict) else {}
def parse_embedding_response(payload: object) -> list[list[float]]: