@irongraph/node-darwin-arm64 0.1.5 → 0.1.7

package/README.md

@@ -1,11 +1,18 @@
 # IronGraph for Node.js
 
-**Node.js access to a GPU-first temporal graph database with built-in streaming and queues.**
+**Node.js access to a CPU graph database with built-in streaming and queues.**
 
-Work with graph relationships, text document and vector search, temporal property history, graph
-analytics, and built-in Streams and Queues. `@irongraph/node` includes native embedded access,
-remote Query API/Bolt clients, and TypeScript declarations. IronGraph is open-source software
-released under the Apache License 2.0.
+Use Cypher to query graph structure, search text and vectors, inspect property history, and
+administer the database. Built-in Kafka-compatible Streams and AMQP-compatible Queues handle event
+and message workloads. `@irongraph/node` provides native embedded access, remote Query API/Bolt
+clients, and TypeScript declarations for Node.js. IronGraph is open-source software released under
+the Apache License 2.0.
+
+## Performance at a glance
+
+Graph queries execute on the CPU against one shared canonical graph. Local text inference can
+use Metal or CUDA independently. Measure complete queries separately from individual graph operations;
+latency depends on the workload, graph size, indexes, and host.
 
 ## Database capabilities
 
@@ -101,9 +108,9 @@ Select projects explicitly with `USE` or the `projectId` argument. Each project
 `KNOWLEDGE`, and `WORKSPACE` layers for source facts, curated facts, and working data. There is
 no implicit default project.
 
-One embedded instance owns its directory; one process selects one device. CPU is the reference
-backend, Metal is the primary local accelerator, and CUDA requires a CUDA-enabled release.
-GPU admission rejects project graphs and indexes that do not fit. IronGraph embeds text
+One embedded instance owns its directory. Graph execution uses one shared CPU graph; text inference
+selects its device independently. CUDA text inference requires a CUDA-enabled release. Parallel reads
+do not wait for writers and may observe mixed values during a multi-record write. IronGraph embeds text
 locally and does not run generative language models.
 
 Use `Client.apiMtls` or `Client.boltMtls` for remote access with certificate, private-key, and