The Hub for AI Coding Models, Tools & Workflows

LiveKit published Python [email protected] on GitHub (2026-09-26; PyPI livekit-agents 1.8.3): Gemini 3.8 Live defaults `tool_behavior=NON_BLOCKING` and sends FunctionTool schemas as `parameters_json_schema` on the text API; OpenAI gains `GPTLiveModel.with_azure` and moves the Realtime inference client into agents core; STT adds AssemblyAI `universal-3-6-pro` plus `language_confidence`, and configurable ElevenLabs realtime chunk duration; voice fixes prevent parallel tools from canceling handoffs, pair tool outputs by call_id, forward frames as-is in fallback/stream adapters, and fix LiveAvatar interruption/plugin registration. Same-day JS `@livekit/[email protected]` deepens agent_turn telemetry and adaptive interruption.

Cline Desktop v0.0.37 (2026-09-26) adds Settings → About (version/channel, check/restart updates, recent release notes, report issue), a one-time What's new dialog after updates, macOS Help→Export Diagnostics, and fixes Bedrock OpenAI inference-profile calls that failed with Unknown parameter reasoningConfig when reasoning effort was set.

Cline Desktop (@cline_dev) has shipped release v0.0.37. The update introduces a comprehensive lifecycle management experience, embedding a dedicated 'About' view within Settings that enables in-app update checks, one-click restarts, and inline changelog inspection. Users are greeted with an updated 'What's New' dialog highlighting recent architectural milestones including SSH remotes, Git worktrees, pull request telemetry, and parallel sub-agents. Crucially, the release patches an AWS Bedrock inference bug that triggered 'Unknown parameter: reasoningConfig' on OpenAI models, while adding a macOS diagnostic export shortcut.

Vercel AI SDK shipped @ai-sdk/[email protected] (2026-09-26), a breaking migration from Sonar Chat Completions to Perplexity Agent API. Replace Sonar model IDs and provider options with Agent API presets, models, and tools; request/response metadata, raw stream events, and usage/cost shapes change; Sonar PDF input and image/video results are unsupported. Follow-up fixes recover terminal text, accept native tool traces (e.g. finance), and dedupe URL citation sources while preserving search-result IDs.

Vercel AI SDK shipped `@ai-sdk/[email protected]` on 2026-09-26: request-level and positioned system-message `reasoningEffortUpdate` now accept `'none'` so GPT-6 Sol/Luna conversations can turn reasoning off mid-thread. Updates are validated against each model’s supported efforts—unsupported request-level updates warn and omit; unsupported historical/message-level updates reject. Companion `@ai-sdk/[email protected]` and `@ai-sdk/[email protected]` pick up the dependency.

Pipecat v1.12.0 (2026-09-26) adds TTSService.pronunciation_transform_ipa() for reusable word→IPA maps across Cartesia/ElevenLabs/Inworld/Deepgram, client-mode MOQTransport reconnect with session-ending markers, Pipecat Classifiers + LLMClassifier, frame-level interruptible, and response_schema on LLM run_inference. PyPI pipecat-ai 1.12.0 is live.

Alibaba’s open-source terminal coding agent Qwen Code shipped v0.24.6 on 2026-09-26 (npm @qwen-code/qwen-code 0.24.6; Desktop desktop-v0.24.6 and SDK TypeScript sdk-typescript-v0.1.16 same day): CLI declares Managed Runtime v2 execute/status/cancel and mounts tool ops on the worker; adds a native Advisor consultation tool with usage limits; `/batch-api` for agent-prepared batch workflows; serve gains guarded Hosted Runtime, Spring control plane + dual-path WebShell, and model/reasoning selection at session creation; core adds Durable Managed Session journal and failover. Release notes list no known breaking changes.

CopilotKit shipped v1.74.0 on 2026-09-25 (npm @copilotkit/react-core 1.74.0): SDKs can consume learned skills from multiple containers, expanding skill assets beyond a single runtime; Web Inspector gains targeted What's New notifications and launcher HUD impression/action tracking. Fixes: threads drawer no longer locks on unresolved entitlements, stop refreshing realtime credentials for sockets that never open, stop per-message state cloning / virtual-scroll tug-of-war in React, and a Memory policy that grants nothing no longer fails the run.

Pydantic AI v2.51.0 (2026-09-25) adds OpenAILiveModel for OpenAI GPT-Live, exposes context_window_used on realtime sessions, tightens Gemini Live id/affective-dialog checks, and raises UserError for forced realtime tool_choice on OpenAI/Azure/xAI. PyPI pydantic-ai 2.51.0 is live.

Claude Code v2.1.283 (2026-09-25) adds availableModelsMatch=exact and deniedModels for managed model allow/deny lists, opt-in x-claude-code-prompt-id gateway hint headers, /doctor prompt-audit for CLAUDE.md/skills/agents/commands, plus Bedrock Mantle upstream and load_test_mode on the Apps Gateway. npm @anthropic-ai/claude-code 2.1.283 is live.

Strands Agents (harness-sdk) python/v1.57.1 (2026-09-25) adds bidi snapshot capture/restore, a vended a2a_client tool, Agent.shutdown() for scope cleanup, CLI Quickstart/Customize/Export replacing Q&A setup plus /tools, and MCP TS client 2.0. PyPI strands-agents 1.57.1 is live.

Google ADK Python v2.10.0 (GitHub 2026-09-25; changelog 2026-09-24) adds experimental skill lifecycle controls (EPHEMERAL, active-skill caps, unload_skill via ADK_ENABLE_SKILL_LIFECYCLE=1), a MongoDB toolset with vector/hybrid search, eval efficiency metrics for duration/tokens/model calls, and better OpenAI reasoning param adaptation plus reasoning-token reporting. PyPI google-adk 2.10.0 is live.

Mem0 Python SDK v2.2.1 (2026-09-25, PyPI mem0ai 2.2.1) makes `Memory`/`AsyncMemory.add()` report only vector-store inserts that actually succeeded—raising `VectorStoreError` when none insert—restores context-manager close semantics, fixes Bedrock Anthropic Converse to read the first text-bearing content block (after optional reasoningContent), applies full Turbopuffer filter operators, and remaps euclidean(_squared)/S3 Vectors scores to `1/(1+distance)`. Same-day Node SDK ts-v3.3.1 stops injecting OpenAI default baseURL/model into non-OpenAI providers.

Prefect shipped FastMCP v4.0.10 (2026-09-25, PyPI fastmcp 4.0.10, “Inside Job”): task-enabled tools remain registered with the task backend even when hidden by Search transforms or CodeMode, and calls via `ctx.fastmcp.call_tool()` (including the search `call_tool` proxy and CodeMode `execute`) now return real results instead of empty task receipts. Also tolerates stdio transport construction failures in `__del__`. Docs at gofastmcp.com.

Comet Opik 2.2.81 (2026-09-25) routes scored entities into annotation queues, optionally provisions an extended free-form SQL ClickHouse user, registers 19 canonical providers so model prices load, and fixes Gemini usage, streamed Mistral reasoning, and slots=True dataclass encoding in the SDK.

OpenHands Agent Canvas v1.24.0 (2026-09-25) adds header workspace-folder toggles, read-only shared automation conversations on cloud, MCP OAuth credentials preserved on cloud saves (skip consent when tokens still work), OpenAI subscription caches scoped by backend, and Node.js ≥24.

Langfuse v4.46.0 (2026-09-25) ships basic skill management (PR #17803) via unstable REST (list/create/get/patch/delete skills, versions, file content), streamlined experiment side-by-side comparison, formatted JSON on dataset items, optional 100-row tracing page size, assistant respect for data-retention, and v3 dataset-run-items reads on replicas.

Cline Desktop v0.0.36 (2026-09-25) skips system HTTP(S) proxies for local hub checks, fits oversized windows to small/scaled displays, loads plugin slash commands on workspace open, and runs Bedrock GPT-6/GPT-5.6 without cross-region inference (India via in. profiles).

DeepSeek AI (@deepseek_ai) has officially launched the standalone desktop application for DeepSeek Harness (dsh) across macOS and Windows. Designed to eliminate the engineering overhead of managing CLI background daemons and local sandbox sockets, DeepSeek Harness Desktop integrates the web UI, the Cordis plugin execution engine, and isolated sandboxing into a native local client. It delivers out-of-the-box support for visual trajectory playback, arbitrary session branching, and multi-model sub-agent delegations.

TRL v1.14.0 (2026-09-25) removes `trl.losses` fused-loss forks: DPO/KTO/GRPO stream via `_ChunkedLogProbFunction` into each trainer's own loss, fixing formula drift and plain-DDP missing all-reduce. Ships an in-tree Triton logprob+entropy kernel (~14× vs selective_log_softmax+entropy_from_logits). On 1×H100, Qwen3-0.6B, bs=4, seq 512: chunked+precompute_ref_log_probs median step 0.1971s (−12.1% vs v1.13 fused) and 4.83GB peak (−31%). Removes six low-use experimental trainers; vLLM window moves to ≤0.30.0.