Backfill issue -177 — 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

◆ Local LLM Runtimes

LocalAI

v3.12.1 NOTES

LocalAI v3.12.1 adds backend tracing and updates llama.cpp for Qwen 3 Coder compatibility.

└──▷ GET THIS VERSION

$ git clone --branch v3.12.1 https://github.com/mudler/LocalAI.git
# already have the repo? check out this version:
$ git checkout v3.12.1

›Adds backend traces via the new traces feature (feat(traces)), enabling observability into backend inference calls.

Was this useful?

llama.cpp

b8123 NOTES

llama.cpp b8123 adds a ROCm 7.2 build target covering ten GPU architectures including gfx942 and gfx1201.

└──▷ GET THIS VERSION

$ git clone --branch b8123 https://github.com/ggml-org/llama.cpp.git
# already have the repo? check out this version:
$ git checkout b8123

›Adds a build target to generate ROCm 7.2 artifacts supporting GPU architectures gfx1151, gfx1150, gfx1200, gfx1201, gfx1100, gfx1101, gfx1030, gfx908, gfx90a, and gfx942.

Was this useful?

◆ AI Model & Data Infrastructure

Ollama

v0.17.0 NOTES

Get up and running with Kimi-K2.6, GLM-5.2, MiniMax, DeepSeek, gpt-oss, Qwen, Gemma and other models.

Ollama v0.17.0 adds ollama launch openclaw for one-command OpenClaw setup with web search and open model support.

└──▷ GET THIS VERSION

$ git clone --branch v0.17.0 https://github.com/ollama/ollama.git
# already have the repo? check out this version:
$ git checkout v0.17.0

└──▷ TRY IT

Spin up a fully configured OpenClaw instance in one step to start using open models like Kimi-K2.5 without manual setup.

$ ollama launch openclaw

›New ollama launch openclaw command installs and configures OpenClaw automatically, enabling use with open models like Kimi-K2.5, GLM-5, and Minimax-M2.5.
›Adds web search capability in OpenClaw when using cloud models, allowing internet-augmented responses.
›macOS and Windows apps now automatically select context length based on available VRAM.

Was this useful?

◆ AI Agent Frameworks

OpenClaw

v2026.2.21 NOTES

Your own personal AI assistant. Any OS.

OpenClaw v2026.2.21 adds Gemini 3.1, Volcano Engine/BytePlus providers, Discord voice channels, subagent threading, and per-channel model overrides.

└──▷ GET THIS VERSION

$ git clone --branch v2026.2.21 https://github.com/openclaw/openclaw.git
# already have the repo? check out this version:
$ git checkout v2026.2.21

└──▷ TRY IT

Join a Discord voice channel for realtime voice conversations using the new /vc command.

$ /vc join

Enable Telegram streaming with the simplified boolean config and set a per-channel model override.

yaml

channels:
  telegram:
    streaming: true
  modelByChannel:
    telegram: google/gemini-3.1-pro-preview

›Adds Gemini 3.1 support via google/gemini-3.1-pro-preview model identifier.
›Adds Volcano Engine (Doubao) and BytePlus as providers/models (including coding variants) with interactive and non-interactive onboarding auth flows keyed to volcengine-api-key.
›Adds per-account/channel defaultTo outbound routing fallback so openclaw agent --deliver can send without requiring an explicit --reply-to when a default target is configured.
›Adds per-channel model overrides via channels.modelByChannel, surfaced in /status output.
›Simplifies Telegram preview streaming config to a single boolean channels.telegram.streaming with auto-mapping of legacy streamMode values.

+12 moreshow less

›Adds Discord stream preview mode for live draft replies with partial/block options and configurable chunking.
›Adds configurable lifecycle status reactions for Discord and Telegram covering queued/thinking/tool/done/error phases with a shared controller and emoji/timing overrides.
›Adds Discord voice channel join, leave, and status via the /vc command, plus auto-join configuration for realtime voice conversations.
›Adds configurable ephemeral defaults for Discord slash-command responses.
›Supports updating Discord forum available_tags via channel edit actions for forum tag management.
›Includes channel topics in trusted inbound metadata on new Discord sessions.
›Adds thread-bound subagent sessions on Discord with per-thread focus/list controls and thread-bound continuation routing for spawned helper agents.
›Defaults subagent spawn depth to maxSpawnDepth=2, enabling depth-1 orchestrator spawning by default.
›Adds owner-ID obfuscation using a dedicated HMAC secret configured via ownerDisplaySecret, decoupled from gateway token handling.
›Switches gateway lock and tool-call synthetic IDs from SHA-1 to SHA-256 for a stronger hash basis while keeping deterministic behavior and lock key format.
›Adds model fallback lifecycle visibility in verbose logs, /status active-model context with fallback reason, and WebUI fallback indicators.
›Adds non-blocking dead-code scans in CI via Knip/ts-prune/ts-unused-exports to surface unused dependencies and exports earlier.

Was this useful?

// END OF TRANSMISSION

The AI Toolchain · curated daily · February 21, 2026
Releases & features from the open-source and commercial AI tools that practitioners actually run.

The AI Toolchain — issue -177, February 21, 2026

LocalAI

llama.cpp

Ollama

OpenClaw