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.
$ git clone --branch v3.35.0 https://github.com/cline/cline.git
# already have the repo? check out this version:$ git checkout v3.35.0
└──▷ HOW TO FIND IT
Enable native tool calling for compatible models to get structured, reliable tool invocations instead of prompt-parsed ones
📍In the Cline extension settings, navigate to Features and enable the "Native Tool Calling" toggle.
Send OpenTelemetry traces to a remote OTLP collector with authentication headers for centralized observability
📍In the Cline extension settings, navigate to the OpenTelemetry section, set the OTLP endpoint URL, and add your authentication headers under "Remote OTEL Settings".
Centrally manage MCP server configuration across a team by adding MCP settings to the remote config schema
📍In the Cline extension settings, navigate to Remote Config and add your MCP server definitions under the MCP settings section (using URL as the MCP server ID).
›Adds native tool calling support, enabling models to use structured tool calls instead of prompt-based tool use
›Adds a native tool call setting with feature flag support for per-user control
›Adds Cerebras GLM 4.6 model support and sets it as the default Cerebras model
›Adds Requesty OAuth with custom base URL support for authenticated Requesty provider connections
›Sets default OpenTelemetry exporters to both console and OTLP, enabling out-of-the-box trace export
+4 moreshow less
›Adds custom header support for remote OpenTelemetry (OTEL) settings
›Adds MCP settings to the remote config, allowing centrally managed MCP server configuration
›Adds support for <think> tags, surfacing chain-of-thought reasoning from compatible models
›Adds domain field to OpenAI-compatible provider telemetry for improved observability
An open-source AI agent that brings the power of Gemini directly into your terminal.
gemini-cli v0.13.0-nightly adds a validate command, extension reload support, model routing for all users, and editable compression threshold.
└──▷ GET THIS VERSION
$ git clone --branch v0.13.0-nightly.20251031.c89bc30d https://github.com/google-gemini/gemini-cli.git
# already have the repo? check out this version:$ git checkout v0.13.0-nightly.20251031.c89bc30d
›Adds validate command for checking CLI configuration or extension correctness.
›Supports reloading extensions (MCP servers) without restarting the CLI session.
›Enables model routing for all users, previously limited to a subset.
›Allows the compression threshold to be edited interactively in the UI.
›Logs extension ID alongside tool call and slash command invocations for better traceability.
└──▷ BREAKING ON UPGRADE
!The default compression threshold has changed, which may alter when context compression is triggered in existing sessions.
LangChain debuts langchain-model-profiles, adding a profile property to BaseChatModel.
└──▷ GET THIS VERSION
$ git clone --branch langchain-model-profiles==0.0.1 https://github.com/langchain-ai/langchain.git
# already have the repo? check out this version:$ git checkout langchain-model-profiles==0.0.1
›Adds a profile property to BaseChatModel via the new langchain-model-profiles package, enabling model metadata profiles to be attached to chat model instances.
LocalAI v3.7.0 adds agentic MCP support with POST /mcp/v1/chat/completions, a neutts TTS backend, /v1/videos endpoint, fuzzy gallery search, and Qwen 3 VL.
└──▷ GET THIS VERSION
$ git clone --branch v3.7.0 https://github.com/mudler/LocalAI.git
# already have the repo? check out this version:$ git checkout v3.7.0
└──▷ USE IT
Configure an MCP-powered agent with DuckDuckGo web search and bounded reasoning loops, then call it via the OpenAI-compatible endpoint.
Generate a short video from a text prompt using the new OpenAI-compatible video endpoint.
$ curl http://localhost:8080/v1/videos \
-H 'Content-Type: application/json' \
-H 'Authorization: Bearer sk-...' \
-d '{"model": "sora", "prompt": "A cat walking through a forest at sunset", "size": "1024x576"}'
Set up the neutts backend for streaming low-latency TTS with a reference audio transcript.
yaml
name: neutts-english
backend: neutts
parameters:
model: neuphonic/neutts-air
tts:
audio_path: ./output.wav
streaming: true
options:
- ref_text: 'So I am live on radio...'
known_usecases:
- tts
›Adds POST /mcp/v1/chat/completions endpoint (OpenAI-compatible) for agentic tool-using AI with MCP, configurable via mcp and agent YAML blocks including max_attempts, max_iterations, enable_reasoning, and enable_re_evaluation keys.
›Adds neutts backend for TTS, configured with backend: neutts in the model YAML, supporting parameters.model, tts.audio_path, tts.streaming, and options (e.g. ref_text) for low-latency natural speech generation.
›Adds POST /v1/videos OpenAI-compatible endpoint for text-to-video generation, accepting model, prompt, and size fields.
›Adds whisper.cpp CPU-variant builds for avx, avx2, avx512, and fallback instruction sets to support a wider range of hardware without crashes.
›Adds Qwen 3 VL model support via llama.cpp/gguf, with gallery entries including qwen3-vl-2b-instruct, qwen3-vl-4b-instruct, qwen3-vl-32b-instruct, qwen3-vl-30b-a3b-instruct, and thinking variants.
+5 moreshow less
›Adds fuzzy and case-insensitive gallery search so queries like gema or gemaa match gemma, gemma-3, gemma2, etc.
›Adds long-form TTS text chunking for the chatterbox backend, splitting long inputs intelligently to preserve audio context.
›Adds strict_mode: true enforcement in tool/function definitions, returning a structured JSON error instead of crashing when a required tool call is invalid.
›Adds a WebUI MCP toggle in the chat interface for models with mcp settings enabled, plus simplified YAML model editing and Alpine.js-powered reactivity replacing HTMX.
›Adds support for nullable JSON schema types (["string", "null"]) and empty parameters in function definitions without panics.
└──▷ BREAKING ON UPGRADE
!Upgrading the llama.cpp backend is required for existing LocalAI installations to gain Qwen 3 VL support and the updated llama.cpp version.
LanceDB v0.22.3-beta.4 updates the Voyage AI embedding integration.
└──▷ GET THIS VERSION
$ git clone --branch v0.22.3-beta.4 https://github.com/lancedb/lancedb.git
# already have the repo? check out this version:$ git checkout v0.22.3-beta.4
›Updates the Voyage AI integration for embeddings.
Updates the VoyageAI embedding integration in LanceDB.
└──▷ GET THIS VERSION
$ git clone --branch python-v0.25.3-beta.4 https://github.com/lancedb/lancedb.git
# already have the repo? check out this version:$ git checkout python-v0.25.3-beta.4
Langfuse v3.124.0 adds @mentions in comments, Authentik SSO, Bedrock cache token parsing, and org API key listing.
└──▷ GET THIS VERSION
$ git clone --branch v3.124.0 https://github.com/langfuse/langfuse.git
# already have the repo? check out this version:$ git checkout v3.124.0
›Adds authentik as a supported auth provider via the feat(auth) integration, enabling Authentik-based SSO for self-hosted deployments.
›Allows listing all organization API keys using org-scoped API keys — expands the org API key permission surface.
›Supports @mentions in comments, enabling direct user tagging within trace and observation comment threads.
›Re-implements the Traces public API v1, updating the underlying execution path for the existing traces endpoint.
›Adds Bedrock cache token parsing in the OpenTelemetry (OTEL) ingestion pipeline, enabling accurate cost and token accounting for Bedrock-cached responses.
+2 moreshow less
›Parses tag.tags properties on incoming OTEL spans, mapping them to Langfuse trace tags.
›Enables multi-instance Docker dev setup configuration via environment variables.
Weave v0.52.14 adds imperative call logging, new CRUD endpoints for Models/EvaluationRuns/Predictions/Scores, and OTEL wb_run_id attribution.
└──▷ GET THIS VERSION
$ git clone --branch v0.52.14 https://github.com/wandb/weave.git
# already have the repo? check out this version:$ git checkout v0.52.14
└──▷ USE IT
Log a call imperatively — useful when you cannot wrap a function with a decorator, such as when tracing third-party code or building a custom instrumentation layer.
THE AI TOOLCHAIN