Backfill issue -062 — 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   # 8 tools matched

AI & LLM Tooling

◆ AI Agent Frameworks

AutoGPT

autogpt-platform-beta-v0.6.64 NOTES

AutoGPT Platform adds AutoPilot context panel, Cmd+K global search, read-only graph builder, and webhook trigger presets in v0.6.64

└──▷ GET THIS VERSION

$ git clone --branch autogpt-platform-beta-v0.6.64 https://github.com/Significant-Gravitas/AutoGPT.git
# already have the repo? check out this version:
$ git checkout autogpt-platform-beta-v0.6.64

›Adds a post_to_discord tool for the copilot bot, enabling proactive outbound delivery to Discord channels and threads via RPC.
›Supports @@agptfile reference expansion inside MCP tool arguments, allowing file-path tokens to be resolved inline.
›New AutoPilot Context Panel (v1) with Shell and Files tabs for in-platform context management.
›Global Cmd+K command palette now available across all platform pages for fast navigation.
›Read-only Builder mode lets users view non-owned graphs in the Builder without edit permissions.

+4 moreshow less

›Copilot Webhook Triggers: webhook-trigger setup with preset lifecycle management for automated agent activation.
›Background 'dream' pass added to the copilot for asynchronous processing.
›Deployable entry point introduced for the BatchExecutor, enabling standalone batch execution infrastructure.
›Adds baseline security response headers and enforces webhook ownership on the ping endpoint.

Was this useful?

◆ AI Coding Agents

OpenAI Codex CLI

rust-v0.141.0 NOTES

Lightweight coding agent that runs in your terminal

Codex CLI gains Noise-encrypted remote execution, per-thread MCP servers, auto-resolving TUI prompts, and richer app-server APIs.

└──▷ GET THIS VERSION

$ git clone --branch rust-v0.141.0 https://github.com/openai/codex.git
# already have the repo? check out this version:
$ git checkout rust-v0.141.0

›Adds authenticated, end-to-end encrypted Noise relay channels for remote executors.
›Cross-platform remote execution now preserves executor-native working directories and shells, including filesystem permission paths across app-server and exec-server boundaries.
›Selected executor plugins can activate their stdio MCP servers per thread; plugin discovery adds a created-by-me marketplace and auth-specific curated catalogs.
›App-server clients can list immediate child threads, correlate external-agent import results, and read or redeem rate-limit reset credits.
›Realtime clients can explicitly append speech, control how Codex responses enter conversations, and omit startup context.

Was this useful?

Earendil Works Pi

v0.79.7 NOTES

AI agent toolkit: unified LLM API, agent loop, TUI, coding agent CLI

Pi v0.79.7 adds automatic light/dark theme switching, Warp inline images, and new extension API helpers.

└──▷ GET THIS VERSION

$ git clone --branch v0.79.7 https://github.com/earendil-works/pi.git
# already have the repo? check out this version:
$ git checkout v0.79.7

└──▷ TRY IT

Update pi and all extensions together in one step, equivalent to the old bare pi update behavior.

$ pi update --all

›Adds automatic theme mode in /settings — choose separate light and dark themes that follow terminal color-scheme changes in real time.
›Adds pi update --all flag; bare pi update now updates pi only, separating pi updates from extension updates.
›Adds inline image rendering for Warp terminal users via Kitty graphics detection.
›Exports edit diff helpers (generateDiffString, generateUnifiedPatch, EditDiffResult) from the public API for extensions that need edit-style diffs.

└──▷ BREAKING ON UPGRADE

!Bare pi update now updates pi only; previously it updated pi and extensions together. Scripts or workflows expecting pi update to also update extensions must switch to pi update --all.
!/ is now reserved in theme names for automatic light/dark theme settings; any existing theme name containing / may conflict with the new automatic mode.

Was this useful?

Google gemini-cli

v0.47.0 NOTES

An open-source AI agent that brings the power of Gemini directly into your terminal.

gemini-cli v0.47.0 adds Antigravity CLI migration commands and promotes browser agent to stable.

└──▷ GET THIS VERSION

$ git clone --branch v0.47.0 https://github.com/google-gemini/gemini-cli.git
# already have the repo? check out this version:
$ git checkout v0.47.0

›Adds documentation and migration commands for the Antigravity CLI.
›Promotes browser agent out of experimental status.
›Updates auto mode to use Gemini 3.5 Flash when the flag is enabled.

Was this useful?

◆ Local LLM Runtimes

llama.cpp

b9699 NOTES

SYCL backend gains MUL_MAT and OUT_PROD support for Q1_0 quantization format

└──▷ GET THIS VERSION

$ git clone --branch b9699 https://github.com/ggml-org/llama.cpp.git
# already have the repo? check out this version:
$ git checkout b9699

›Adds MUL_MAT and OUT_PROD operation support for the Q1_0 quantization format in the SYCL backend.

Was this useful?

vMLX

v1.5.65 3 RELEASES · 2026-06-18 NOTES STABLE

vMLX - JANGTQ Uber Compressed MLX Models - L2 Disk Cache (survives restart) + L1 Paged (super fast ttft) + Hybrid SSM Scheduler + Cont Batching + etc!

vMLX 1.5.65 adds MiniMax-M3 and Gemma 4 VL compatibility with reasoning mode, VL image input, and multi-API support.

└──▷ GET THIS VERSION

$ git clone --branch v1.5.65 https://github.com/jjang-ai/vmlx.git
# already have the repo? check out this version:
$ git checkout v1.5.65

›Adds MiniMax-M3 (MM3) model support with startup configuration: prefix cache enabled, paged cache disabled, disk prompt cache enabled, JIT disabled, native MSA/Lightning sparse cache storage, and generic TurboQuant KV disabled.
›Supports MM3 reasoning mode in three states: off, on, and auto, across UI and API paths.
›Supports visible content emission and exact tool calls for MM3 across UI and API paths.
›Extends API compatibility to cover Responses, Chat Completions, Anthropic-compatible, and Ollama-compatible endpoints for MM3.
›Adds streaming support for MM3 across all covered API paths.

+3 moreshow less

›Adds VL image input support for MM3.
›Adds prefix-cache reuse telemetry for MM3.
›Updates latest.json to point to the notarized 1.5.65 Sequoia and Tahoe DMGs.

2 more releases in this issue · 2026-06-18

v1.5.64 NOTES STABLE

vMLX 1.5.64 adds MiniMax-M3 and Gemma 4 VL model support with streaming reasoning fixes and cache configuration defaults.

└──▷ GET THIS VERSION

$ git clone --branch v1.5.64 https://github.com/jjang-ai/vmlx.git
# already have the repo? check out this version:
$ git checkout v1.5.64

›Adds MiniMax-M3 (MM3) model support across live UI, Chat Completions, Responses, Anthropic-compatible, Ollama-compatible, and streaming tool/image paths.
›MiniMax-M3 startup defaults: prefix cache enabled, paged cache disabled, disk prompt cache enabled, JIT disabled, native MSA/Lightning sparse cache storage enabled, generic TurboQuant KV disabled.
›Adds Gemma 4 E2B, E4B, 12B, 26B, and 31B JANG_4M VL variants with live UI, API, and cache coverage, plus 26B and 31B MXFP4 visual row support.
›Streaming reasoning-only responses now produce a bounded visible-answer pass for supported reasoning models, preventing blank output when the model emits hidden reasoning only.
›Updates latest.json to point to the notarized 1.5.64 Sequoia and Tahoe DMGs, which are Developer ID signed, Apple notarized, stapled, and Gatekeeper accepted.

v1.5.63 NOTES STABLE

vMLX v1.5.63 adds MiniMax-M3 and Gemma 4 E2B/E4B/12B MXFP4 vision/text compatibility with multimodal routing and reasoning controls.

└──▷ GET THIS VERSION

$ git clone --branch v1.5.63 https://github.com/jjang-ai/vmlx.git
# already have the repo? check out this version:
$ git checkout v1.5.63

›Adds MiniMax-M3 compatibility with multimodal routing, reasoning controls (off/on/auto), tool calls, and prefix-cache hit stability across Chat/Responses/Anthropic/Ollama endpoints.
›Adds Gemma 4 E2B, E4B, and 12B MXFP4 vision/text compatibility with reasoning controls, image input, tool calls, and cache hits across Chat/Responses/Anthropic/Ollama and streaming surfaces.
›Preserves model-owned generation defaults for supported startup paths.
›MiniMax-M3 enforces its native MSA/Lightning sparse cache policy: paged cache disabled, TurboQuant KV skipped for native MSA, and JIT disabled for the dynamic sparse-cache path.

Was this useful?

Other / Uncategorized

◆ VECTOR DB RAG

LanceDB

v0.31.0-beta.0 2 RELEASES · 2026-06-18 NOTES STABLE

LanceDB v0.31.0-beta.0 adds table branches, FM-Index substring search, approx vector query mode, and richer IndexConfig metadata.

└──▷ GET THIS VERSION

$ git clone --branch v0.31.0-beta.0 https://github.com/lancedb/lancedb.git
# already have the repo? check out this version:
$ git checkout v0.31.0-beta.0

›Adds FM-Index scalar index type for substring search via feat: support FM-Index scalar index for substring search.
›Adds approx mode to vector queries, enabling approximate nearest-neighbor search control.
›Adds isin support to the Expr builder for set-membership filter expressions.
›Accepts Expr in Table.delete and in merge when_not_matched_by_source_delete (Python).
›Expands IndexConfig with rich per-index metadata, now exposed in Python and Node.js bindings.

+5 moreshow less

›Adds table branch support, including checkout of a specific version on a branch, for remote tables and Python/TypeScript bindings.
›Implements set/unset_lsm_write_spec as a REST variant for remote tables.
›Supports rename_table on LanceNamespaceDatabase.
›Adds connect and update column metadata capabilities.
›Sends a read-freshness signal on the lance-namespace path.

└──▷ BREAKING ON UPGRADE

!The loss field is removed from IndexStatistics (dropped as unused).

1 more release in this issue · 2026-06-18

python-v0.34.0-beta.0 NOTES STABLE

LanceDB python-v0.34.0-beta.0 adds FM-Index substring search, table branching, approx vector query mode, and richer IndexConfig metadata.

└──▷ GET THIS VERSION

$ git clone --branch python-v0.34.0-beta.0 https://github.com/lancedb/lancedb.git
# already have the repo? check out this version:
$ git checkout python-v0.34.0-beta.0

└──▷ USE IT

Filter rows using set-membership in a query expression instead of hand-crafting a SQL IN clause.

python

from lancedb.query import Expr
results = table.search().where(Expr.col("status").isin(["active", "pending"])).to_list()

›Adds FM-Index scalar index type for substring search, enabling efficient sub-string queries on text columns.
›Adds isin support to the Expr builder for set-membership filtering.
›Accepts Expr objects in Table.delete and in merge when_not_matched_by_source_delete, replacing raw SQL strings.
›Adds approx mode to vector queries, letting callers trade recall for speed at query time.
›Adds table branch support to local tables, remote tables, and Python/TypeScript bindings, including the ability to check out a specific version on a branch.

+5 moreshow less

›Expands IndexConfig with rich per-index metadata fields, exposed in both Python and Node.js bindings.
›Implements set/unset LSM write spec via the REST remote variant.
›Adds rename_table support on LanceNamespaceDatabase.
›Adds column metadata connect and update capabilities.
›Sends a read-freshness signal on the lance-namespace path.

└──▷ BREAKING ON UPGRADE

!The loss field is removed from IndexStatistics (dropped as unused).

Was this useful?

◆ AI OBSERVABILITY

Langfuse

v3.192.2 5 RELEASES · 2026-06-18 NOTES STABLE

Langfuse v3.192.2 adds BullMQ stall and blob export observability to the worker.

└──▷ GET THIS VERSION

$ git clone --branch v3.192.2 https://github.com/langfuse/langfuse.git
# already have the repo? check out this version:
$ git checkout v3.192.2

›Adds BullMQ stall and blob export observability to the worker service for monitoring job health and export operations.

4 more releases in this issue · 2026-06-18

v3.192.1 NOTES STABLE

Langfuse v3.192.1 adds an API endpoint to invalidate all cached API keys at once.

└──▷ GET THIS VERSION

$ git clone --branch v3.192.1 https://github.com/langfuse/langfuse.git
# already have the repo? check out this version:
$ git checkout v3.192.1

›Adds an API endpoint to invalidate all cached API keys in bulk.

v3.192.0 NOTES STABLE

Langfuse v3.192.0 adds a legacy public API rate limit group, agent system prompt loading from Langfuse, and S3 signing diagnostics on upload failures.

└──▷ GET THIS VERSION

$ git clone --branch v3.192.0 https://github.com/langfuse/langfuse.git
# already have the repo? check out this version:
$ git checkout v3.192.0

›Adds a legacy public API rate limit group to manage rate limiting for older API consumers separately.
›Enables agents to load their system prompt directly from Langfuse, centralizing prompt management.
›Logs S3 signing diagnostics on upload failures to aid in debugging blob storage configuration issues.

v3.191.0 NOTES STABLE

Langfuse v3.191.0 makes grammar search GA on v4 events tables and adds timing to the blob export pipeline.

└──▷ GET THIS VERSION

$ git clone --branch v3.191.0 https://github.com/langfuse/langfuse.git
# already have the repo? check out this version:
$ git checkout v3.191.0

›Promotes the grammar search bar to general availability on the v4 events tables.
›Adds application-level timing instrumentation to the blob export pipeline for observability into export performance.

v3.189.0 NOTES STABLE

Langfuse v3.189.0 adds grammar search for the v4 events table, Flue framework OTEL span mapping, and a codebase navigator agent skill.

└──▷ GET THIS VERSION

$ git clone --branch v3.189.0 https://github.com/langfuse/langfuse.git
# already have the repo? check out this version:
$ git checkout v3.189.0

›Adds grammar search bar (beta) to the v4 events table in the observations UI, enabling structured filtering of trace data.
›Maps Flue framework spans to the Langfuse data model via OpenTelemetry, extending OTEL ingestion support to the Flue framework.
›Adds a Langfuse codebase navigator skill to the agents feature, enabling AI-assisted exploration of the Langfuse codebase.
›Adds a redirect tool to the agent framework for handling agent-driven redirects.
›Lands new users in a default project automatically during onboarding, reducing initial setup friction.

+1 moreshow less

›Adds an overlay layer system to the UI for improved rendering of layered components such as search-bar tooltips.

Was this useful?

// END OF TRANSMISSION

The AI Toolchain · curated daily · June 18, 2026
Releases & features from the open-source and commercial AI tools that practitioners actually run.

The AI Toolchain — issue -062, June 18, 2026