Backfill issue -434 — The AI Toolchain

EDITIONS tail grep head diff uniq

The daily firehose — everything the toolchain shipped today, already filtered.

// HOW THIS ISSUE IS MADE

We read every release from the 174 tools on our watchlist at the source — GitHub and GitLab release notes, vendor release pages and changelogs, project blogs and feeds, vendor press releases, and the source code behind the tag. Bug-fix-only releases and non-product newsroom noise are dropped; what's left is summarized down to the new capability, how to try it, and any screenshots or videos the release itself published. Every entry links to the sources it was built from.

VIEW

ISSUE VIEW full issue

Do you prefer this view?

$ tct list   # 2 tools matched

AI & LLM Tooling

◆ AI Model & Data Infrastructure

NVIDIA Triton Inference Server

v1.6.0 2 RELEASES · 2019-09-03 → 2019-09-27 NOTES STABLE

Triton v1.6.0 adds TensorRT 6 dynamic shapes, shared memory tensors, S3 model repos, and a new library API.

└──▷ GET THIS VERSION

$ git clone --branch v1.6.0 https://github.com/triton-inference-server/server.git
# already have the repo? check out this version:
$ git checkout v1.6.0

›Supports Amazon S3 as a remote model repository via the s3:// prefix on model repository paths.
›Adds shared memory support (alpha) so input and output tensors can be communicated via system (CPU) shared memory instead of over the network, reducing data-copy overhead.
›Introduces a libtrtserver.so library API (beta) that lets applications link the full inference server functionality directly instead of communicating over HTTP/gRPC.
›Adds TensorRT 6 support, including dynamic shapes.
›Extends perf_client to support models with variable-sized input tensors.

+3 moreshow less

›The gRPC endpoint now uses significantly less memory while delivering higher throughput.
›The ensemble scheduler now allows batching and non-batching models to be composed together.
›The ensemble scheduler retains tensors in GPU memory between models when possible, avoiding round-trips through system memory.

1 more release in this issue · 2019-09-03 → 2019-09-27

v1.5.0 NOTES STABLE

Triton v1.5.0 adds a GRPC/HTTP model control API, multi-GPU TF distribution, and a C++ custom backend SDK wrapper.

└──▷ GET THIS VERSION

$ git clone --branch v1.5.0 https://github.com/triton-inference-server/server.git
# already have the repo? check out this version:
$ git checkout v1.5.0

›Adds a new GRPC/HTTP model control API for dynamically loading and unloading models at runtime, enabling an execution mode where the server starts with no models loaded from the model repository.
›Adds a new instance-group mode for TensorFlow models to explicitly distribute inferencing across multiple GPUs within the inference server.
›Extends perf_client to report per-model statistics and aggregate ensemble statistics for ensemble models.
›Adds a C++ wrapper around the custom backend C API in the custom backend SDK to simplify authoring custom backends.
›Improves input/output tensor reshape to support variable-sized dimensions.

Was this useful?

Other / Uncategorized

◆ VECTOR DB RAG

Weaviate

0.20.0 NOTES

Weaviate 0.20.0 replaces Janusgraph/Cassandra with an esvector-only backend, merges the GraphQL Meta and Aggregate APIs, and adds forced index refresh on missing cross-references.

└──▷ GET THIS VERSION

$ git clone --branch 0.20.0 https://github.com/weaviate/weaviate.git
# already have the repo? check out this version:
$ git checkout 0.20.0

└──▷ USE IT

Tune how many cross-reference levels Weaviate caches to balance query depth against storage cost — increase for deep, narrow schemas; decrease for shallow, wide ones.

yaml

vector_index:
  denormalizationDepth: 4

›Replaces the Janusgraph + Elasticsearch + Cassandra stack with a single vector-optimized Elasticsearch ('esvector') backend, delivering faster listing queries, lower infrastructure footprint, and fully integrated native vector search.
›Configures cross-reference denormalization depth via vector_index.denormalizationDepth in config.yaml (default: 3), controlling how many reference levels are cached in the background for efficient traversal and filtering.
›Forces an Elasticsearch index refresh when a cross-referenced object is not yet visible on the index, then retries immediately — eliminating the need for client-side retry logic when adding objects with cross-references in rapid succession.
›GraphQL Meta API is merged into the Aggregate API, with grouping now an optional parameter rather than always on or always off.
›Distinguishes text properties (mapped as Elasticsearch text, for full-text fields) from string properties (mapped as Elasticsearch keyword, for exact values like emails and IDs), with aggregations now supported only on string props.

└──▷ BREAKING ON UPGRADE

!The GraphQL Meta API is merged into the Aggregate API; any queries targeting the separate Meta API will break.
!The base unit for geoCoordinates search distance changed from kilometer to meter; existing query values must be multiplied by 1000.
!Aggregations (e.g. top-N value counts) on text properties are no longer supported; only string properties support aggregations from 0.20.0 onward.

Was this useful?

// END OF TRANSMISSION

The AI Toolchain · curated daily · September 27, 2019
Releases & features from the open-source and commercial AI tools that practitioners actually run.

The AI Toolchain — issue -434, September 27, 2019

NVIDIA Triton Inference Server

Weaviate