Documentation Index
Fetch the complete documentation index at: https://docs.9dlabs.xyz/llms.txt
Use this file to discover all available pages before exploring further.
The v2 client is available on Pro, Team, and Enterprise plans.
Quick Start
from nined.memory import MemoryClientV2
client = MemoryClientV2(
base_url="https://api.9dlabs.xyz",
api_key="your-key",
workspace_id="my-workspace",
)
# Batch ingest
client.ingest([
{"artifact_type": "document", "raw_payload": {"content": "Deploy runbook..."}},
{"artifact_type": "note", "raw_payload": {"content": "Meeting notes..."}},
])
# Retrieve with serving profile
pack = client.context_pack("What happened?", profile="high_recall")
for snippet in pack["snippets"]:
print(snippet["content"])
Constructor
MemoryClientV2(
base_url: str = "https://api.9dlabs.xyz",
api_key: str = "",
workspace_id: str = "",
timeout: int = 30,
)
| Parameter | Type | Default | Description |
|---|
base_url | str | https://api.9dlabs.xyz | Server URL |
api_key | str | "" | API key |
workspace_id | str | "" | Default workspace (used for all methods) |
timeout | int | 30 | HTTP timeout in seconds |
Key Differences from v1
| Feature | v1 | v2 |
|---|
| Ingest | Single artifact | Batch (array) |
| Workspace | Per-method parameter | Set once on client |
| Serving profiles | Not available | low_latency, balanced, high_recall |
| Async indexing | Via env var | Per-request via async_index param |
| Job tracking | Not available | job_status() method |
Methods
ingest()
Batch ingest multiple artifacts. Supports async indexing with job tracking.
# Synchronous
result = client.ingest([
{"artifact_type": "document", "raw_payload": {"content": "..."}},
{"artifact_type": "chat_turn", "raw_payload": {"role": "user", "content": "..."}},
])
# Async with job tracking
result = client.ingest([...], async_index=True)
for job in result["queued_jobs"]:
status = client.job_status(job["job_id"])
print(f"Job {job['job_id']}: {status['status']}")
context_pack()
Retrieve context with serving profiles for different latency/recall tradeoffs.
pack = client.context_pack(
"What is our pricing policy?",
max_tokens=8192,
profile="high_recall",
)
job_status()
Track async indexing jobs.
status = client.job_status("job-uuid")
# Returns: {"job_id": "...", "status": "done", "artifact_id": "...", ...}
ready()
Readiness check including storage health.
health = client.ready()
# Returns: {"status": "ok", "storage": "connected", ...}
Other Methods
The following methods work the same as v1 but use the workspace set on the client:
feedback(action, artifact_id, ...) — Submit correctionlist_receipts(limit, offset) — List receiptsget_receipt(pack_hash) — Get receipt by pack hashartifact_status(artifact_id) — Indexing statusask(query, max_tokens, profile) — LLM-synthesized answerdelete_workspace() — Delete workspacehealth() — Liveness check
Serving Profiles
| Profile | Use Case | Tradeoff |
|---|
low_latency | Real-time chat, quick lookups | Fastest retrieval, fewer results |
balanced | General-purpose (default) | Good recall with reasonable latency |
high_recall | Research, compliance, audits | Maximum evidence, higher latency |
