Backfill issue -046 — 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 Coding Agents

Cline

cli-v3.0.37 NOTES

Autonomous coding agent as an SDK, IDE extension, or CLI assistant.

Cline v3.0.37 adds graceful handling of malformed tool calls and exposes plan/act mode switches to the model.

└──▷ GET THIS VERSION

$ git clone --branch cli-v3.0.37 https://github.com/cline/cline.git
# already have the repo? check out this version:
$ git checkout cli-v3.0.37

›Handles malformed tool calls from weaker models (e.g. DeepSeek) — wrong argument types or truncated JSON — gracefully, running instead of erroring out.
›Plan/act mode switches are now visible to the model, so it is aware when you change modes mid-session.

Was this useful?

◆ Local LLM Runtimes

LocalAI

v4.6.0 NOTES

LocalAI v4.6.0 adds a load API, PII Prometheus counter, chat forking, and AMD ROCm GPU fixes

└──▷ GET THIS VERSION

$ git clone --branch v4.6.0 https://github.com/mudler/LocalAI.git
# already have the repo? check out this version:
$ git checkout v4.6.0

└──▷ TRY IT

Pre-warm a specific model into VRAM before the first request arrives, avoiding cold-start latency for users.

$ curl -X POST http://localhost:8080/backend/load -H 'Content-Type: application/json' -d '{"model": "mistral-7b"}'

Keep a realtime pipeline from blocking on warm-up for a low-latency session where cold-start is acceptable.

yaml

disable_warmup: true

›Adds POST /backend/load and POST /v1/backend/load endpoints to pre-warm any model (including full pipelines) on demand, complementing the existing /backend/shutdown.
›Adds disable_warmup: true per-pipeline config key to opt out of eager realtime session warm-up.
›Adds LOCALAI_BACKEND_PARENT_WATCH environment variable (default on, auto-off on Windows) to make spawned backend workers self-terminate when the parent LocalAI process dies.
›Adds LOCALAI_BACKEND_PARENT_WATCH_INTERVAL environment variable (default 2s) to control how frequently backend workers poll for parent-process death.
›Adds "force": true field to POST /backends/apply to force re-download of an already-installed backend; installs are now idempotent by default.

+7 moreshow less

›Bundles HIPBLASLT_TENSILE_LIBPATH export in llama-cpp and turboquant run.sh, enabling full-speed hipBLASLt kernel-tuning data for AMD ROCm backends.
›Adds a load_model MCP admin tool and 'Load into memory' UI button, both backed by the new POST /backend/load endpoint.
›Adds LOCALAI_EXTERNAL_BACKENDS boot loop idempotency so backends are not re-pulled on every restart unless force: true is passed.
›Adds conversation-forking to the built-in React chat UI: regenerate any assistant answer, branch a new chat from any turn, duplicate a chat, or copy the whole conversation as Markdown.
›Realtime voice sessions now eagerly warm the full pipeline (VAD, ASR, LLM, TTS) at session start, with concurrent sub-model loading so warm-up time equals the slowest stage, not the sum.
›AMD ROCm rocm-vllm backend now installs from the ROCm wheel index (https://wheels.vllm.ai/rocm/) pinned to CPython 3.12, replacing the CUDA-only PyPI wheel.
›AMD ROCm ggml audio backends (rocm-qwen3-tts-cpp, rocm-omnivoice-cpp, acestep-cpp, vibevoice-cpp) now build with -DGGML_HIP=ON and link the HIP backend for real GPU offload.

Was this useful?

◆ AI Agent Frameworks

PydanticAI

v2.5.0 NOTES

PydanticAI v2.5.0 adds sanitize_messages for message-history hardening and multimodal tool-return round-trips in AG-UI and Vercel AI adapters.

└──▷ GET THIS VERSION

$ git clone --branch v2.5.0 https://github.com/pydantic/pydantic-ai.git
# already have the repo? check out this version:
$ git checkout v2.5.0

›Adds sanitize_messages for inbound message-history hardening, closing a dangling-tool-call re-exposure on the Agent.to_ag_ui() / AGUIAdapter serving path.
›Supports round-trip multimodal tool returns through the AG-UI and Vercel AI adapters, covering both history and streaming paths.

Was this useful?

Other / Uncategorized

◆ AI OBSERVABILITY

Arize Phoenix

arize-phoenix-v17.17.0 2 RELEASES · 2026-07-04 NOTES STABLE

Phoenix v17.17.0 adds shift-click range selection in tables and recent searches in the span filter typeahead.

└──▷ GET THIS VERSION

$ git clone --branch arize-phoenix-v17.17.0 https://github.com/Arize-ai/phoenix.git
# already have the repo? check out this version:
$ git checkout arize-phoenix-v17.17.0

›Adds recent search history to the span filter typeahead for faster reuse of previous filter queries.
›Enables shift-click range selection in tables for bulk-selecting multiple rows at once.

1 more release in this issue · 2026-07-04

arize-phoenix-v17.16.0 NOTES STABLE

Phoenix v17.16.0 adds REST endpoints for annotation configs and dataset labels, CLI annotation-config commands, and JS precision/recall/F-score evaluators.

└──▷ GET THIS VERSION

$ git clone --branch arize-phoenix-v17.16.0 https://github.com/Arize-ai/phoenix.git
# already have the repo? check out this version:
$ git checkout arize-phoenix-v17.16.0

└──▷ TRY IT

Filter projects by name substring when listing all projects — useful when managing large numbers of tracing projects.

$ curl -X GET 'https://<phoenix-host>/v1/projects?name_contains=prod' -H 'Authorization: Bearer <token>'

›Adds name_contains filter parameter to GET /v1/projects for server-side project name filtering.
›Adds REST endpoints for assigning annotation configs to projects via the API.
›Adds dataset label REST endpoints to the server.
›Adds annotation-config get, annotation-config create, and annotation-config update subcommands to the CLI.
›Adds precision, recall, and F-score code evaluators to the JavaScript phoenix-evals package.

+5 moreshow less

›Adds auth.md support for agent authentication discovery.
›Adds baseline tagging for experiments.
›Adds an error column to the spans and traces tables in the UI.
›Adds version count, latest version, and version tags columns to the prompts table in the UI.
›Adds a dismissable version update notice to the side navigation in the UI.

Was this useful?

// END OF TRANSMISSION

The AI Toolchain · curated daily · July 4, 2026
Releases & features from the open-source and commercial AI tools that practitioners actually run.

The AI Toolchain — issue -046, July 4, 2026

Cline

LocalAI

PydanticAI

Arize Phoenix