Backfill issue -402 — 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   # 4 tools matched

AI & LLM Tooling

◆ AI Agent Frameworks

deepset Haystack

v1.4.0 NOTES

Haystack v1.4.0 adds MLflow eval tracking, FARMReader confidence filtering, Milvus2 vector+metadata queries, and BM25Retriever rename.

└──▷ GET THIS VERSION

$ git clone --branch v1.4.0 https://github.com/deepset-ai/haystack.git
# already have the repo? check out this version:
$ git checkout v1.4.0

└──▷ USE IT

Log and compare eval results from multiple pipeline configurations to an MLflow tracking server.

python

eval_result = Pipeline.execute_eval_run(
    index_pipeline=index_pipeline,
    query_pipeline=query_pipeline,
    evaluation_set_labels=labels,
    corpus_file_paths=file_paths,
    corpus_file_metas=file_metas,
    experiment_tracking_tool="mlflow",
    experiment_tracking_uri="http://localhost:5000",
    experiment_name="my-query-pipeline-experiment",
    experiment_run_name="run_1",
    pipeline_meta={"name": "my-pipeline-1"},
    evaluation_set_meta={"name": "my-evalset"},
    corpus_meta={"name": "my-corpus"},
    add_isolated_node_eval=True,
    reuse_index=False
)

Filter out low-confidence reader predictions to reduce noise in QA pipeline answers.

python

from haystack.nodes import FARMReader
model = "deepset/roberta-base-squad2"
reader = FARMReader(model, confidence_threshold=0.5)

›Adds MLflowTrackingHead and Pipeline.execute_eval_run() method with parameters experiment_tracking_tool, experiment_tracking_uri, experiment_name, experiment_run_name, pipeline_meta, evaluation_set_meta, corpus_meta, add_isolated_node_eval, and reuse_index to log evaluation metrics and pipeline artifacts to MLflow.
›Adds confidence_threshold parameter to FARMReader (float between 0 and 1, disabled by default) to filter out low-confidence predictions at initialization time.
›Adds devices parameter alongside existing use_gpu in FARMReader for explicit device assignment.
›Adds alias support in ElasticsearchDocumentStore for querying via index aliases.
›Adds conjunctive query support in sparse retrieval.

+6 moreshow less

›Adds a flag to disable scaling scores to probabilities in retrieval.
›Introduces Milvus2DocumentStore (superseding the now-deprecated Milvus1DocumentStore) with support for filtering by scalar data types alongside vector similarity queries.
›Renames ElasticsearchRetriever to BM25Retriever and ElasticsearchFilterOnlyRetriever to FilterRetriever; deprecated names remain functional until a future release.
›Adds EvaluationSetClient for deepset Cloud to fetch evaluation sets.
›Adds table linearization support in EmbeddingRetriever for table inputs.
›Adds file content-based extension detection (extracts extension based on file content rather than filename).

└──▷ BREAKING ON UPGRADE

!Return types of indexing pipeline nodes have changed.
!weaviate-client is upgraded to 3.3.3, which may affect existing Weaviate integrations.
!TransformersReader defaults are now aligned with FARMReader, changing previous default behavior.
!Default encoding for PDFToTextConverter changed from Latin 1 to UTF-8.
!YAML files are now validated without loading nodes, changing pipeline validation behavior.

Was this useful?

◆ AI Model & Data Infrastructure

NVIDIA Triton Inference Server

v2.22.0 NOTES

Triton v2.22.0 adds Java in-process API, BF16 support, decoupled Python backend, and dynamic model repository registration.

└──▷ GET THIS VERSION

$ git clone --branch v2.22.0 https://github.com/triton-inference-server/server.git
# already have the repo? check out this version:
$ git checkout v2.22.0

›Adds Java bindings for the Triton In-Process API, enabling the full in-process server to be embedded in Java applications.
›Supports loading models with file content supplied inline at API invocation time via the model-repository load extension.
›Adds BF16 (bfloat16) as a supported data type in model configuration.
›Adds support for customized gRPC channel settings in the Python gRPC client library (tritonclient.grpc).
›Python backend adds decoupled API support (BETA), enabling models to send multiple, asynchronous responses per request.

+3 moreshow less

›Explicit model control mode now supports loading all models at startup.
›PyTorch backend adds support for 1-dimensional String tensor I/O.
›ONNX Runtime backend updated to version 1.11.1 on both Ubuntu and Windows.

Was this useful?

◆ AI Coding Agents

Zed

v0.35.0 4 RELEASES · 2022-05-05 → 2022-05-24 NOTES STABLE

Zed v0.35.0 adds new color themes, git-ignored file visibility, a richer app menu, and a collaboration invite flow.

└──▷ GET THIS VERSION

$ git clone --branch v0.35.0 https://github.com/zed-industries/zed.git
# already have the repo? check out this version:
$ git checkout v0.35.0

›Adds multiple new color themes.
›Git-ignored files and directories are now displayed in the project panel.
›Expands the application menu into a more full-featured menu bar.
›Adds a flow for inviting new users to collaborate, with invitations being allocated to insiders.

3 more releases in this issue · 2022-05-05 → 2022-05-24

v0.34.0 NOTES STABLE

Zed v0.34.0 removes explicit project sharing — any contact can now request to join your project directly.

└──▷ GET THIS VERSION

$ git clone --branch v0.34.0 https://github.com/zed-industries/zed.git
# already have the repo? check out this version:
$ git checkout v0.34.0

›Enables join requests for any contact's project without requiring the host to explicitly share first; the host receives a notification and can accept or decline.

v0.33.0 NOTES STABLE

Zed v0.33.0 adds a contact search modal for finding and inviting collaborators.

└──▷ GET THIS VERSION

$ git clone --branch v0.33.0 https://github.com/zed-industries/zed.git
# already have the repo? check out this version:
$ git checkout v0.33.0

›Adds a contact search modal, opened via a button at the top of the contacts panel, to find and invite other Zed users for collaboration.
›Notifies users when a contact request is received, enabling an approval-based collaboration workflow.
›Once a contact relationship is established, contacts can freely join any project the user is sharing.

v0.32.0 NOTES STABLE

Zed v0.32.0 adds project panel file management and a new transpose command bound to ctrl-t.

└──▷ GET THIS VERSION

$ git clone --branch v0.32.0 https://github.com/zed-industries/zed.git
# already have the repo? check out this version:
$ git checkout v0.32.0

›Adds transpose command bound to ctrl-t that swaps the characters before and after the cursor.
›Adds the ability to add, rename, and delete files and folders directly from the project panel, for both local and remote projects.
›Adds missing syntax highlights to the base16 themes (solarized, cave, and sulphurpool).

Was this useful?

Other / Uncategorized

◆ VECTOR DB RAG

Weaviate

v1.13.2 2 RELEASES · 2022-05-03 → 2022-05-20 NOTES STABLE

Weaviate v1.13.2 previews L2 distance support, with full availability planned for v1.14.0.

└──▷ GET THIS VERSION

$ git clone --branch v1.13.2 https://github.com/weaviate/weaviate.git
# already have the repo? check out this version:
$ git checkout v1.13.2

›Adds L2 distance metric support (preview/experimental — full support coming in v1.14.0).

1 more release in this issue · 2022-05-03 → 2022-05-20

v1.13.0 NOTES STABLE

Weaviate v1.13.0 adds faceted vector search, result sorting, timestamp filtering, batch delete by filter, and DPR transformer support.

└──▷ GET THIS VERSION

$ git clone --branch v1.13.0 https://github.com/weaviate/weaviate.git
# already have the repo? check out this version:
$ git checkout v1.13.0

└──▷ TRY IT

Filter objects by creation timestamp — after enabling timestamp indexing — to scope queries to recently ingested data.

$ {
  Get {
    Article(
      where: {
        path: ["_creationTimeUnix"]
        operator: GreaterThan
        valueString: "1672531200000"
      }
    ) {
      title
    }
  }
}

Run a faceted vector search by combining nearText with Aggregate to count matching classes within a vector-search radius.

$ {
  Aggregate {
    Article(
      nearText: {
        concepts: ["machine learning"]
        certainty: 0.75
      }
    ) {
      meta { count }
      category { groupedBy { value } count }
    }
  }
}

›Adds path: ["_creationTimeUnix"] and path: ["_lastUpdateTimeUnix"] filter notation after optionally including creationTimeUnix and lastUpdateTimeUnix in the inverted index — enabling timestamp-based filtering for the first time.
›Adds a new /v1/batch endpoint supporting delete-by-filter, removing all objects that match a specified filter in one operation.
›Enables combining nearVector, nearObject, nearText, and other near<Media> vector searches with Aggregate queries for faceted vector search; requires an explicit limit or a certainty/distance threshold.
›Adds sorting of search results (reads affected objects from disk; columnar-storage optimization planned for a future release).
›Supports DPR (Dense Passage Retrieval) transformer models in text2vec-transformers, using two separate models to encode queries and passages independently.

Was this useful?

// END OF TRANSMISSION

The AI Toolchain · curated daily · May 26, 2022
Releases & features from the open-source and commercial AI tools that practitioners actually run.

The AI Toolchain — issue -402, May 26, 2022

deepset Haystack

NVIDIA Triton Inference Server

Zed

Weaviate