⚡Trending:
H
Hugging Facehuggingface·37m ago
🚀 Release

Hugging Face Transformers 5.18.0: Nemotron 3 streaming diarization; NemotronH Omni; HyperCLOVAX Vision V2; GTE

Hugging Face shipped [`transformers` `v5.18.0`](https://github.com/huggingface/transformers/releases/tag/v5.18.0) on 2026-09-30 (PyPI [`5.18.0`](https://pypi.org/project/transformers/5.18.0/)): NVIDIA **Nemotron 3 Diarization** (streaming/offline, up to 8 speakers, 80 ms–30.4 s buffers), **NemotronH Omni** (joint text/image/video/audio), NAVER **HyperCLOVAX Vision V2** (built-in `<think>` CoT), and **GTE** bidirectional encoders (RoPE + gated MLP; gte / snowflake-arctic-embed family). Official model_doc pages are live.

Hugging Face Transformers 5.18.0: Nemotron 3 streaming diarization; NemotronH Omni; HyperCLOVAX Vision V2; GTE
⚡ Key Takeaways
  • •Release v5.18.0 at 2026-09-30T16:46:27Z; PyPI ~2026-09-30T16:45:57Z
  • •Nemotron 3 Diarization: streaming+offline, ≤8 speakers, 80 ms–30.4 s buffers, 10 ms frame multiples (#49056)
  • •NemotronH Omni: NemotronH + RADIO + optional Parakeet for joint text/image/video/audio (#46509)
  • •HyperCLOVAX Vision V2 (NAVER) + GTE encoders (RoPE/gated MLP; gte / snowflake-arctic-embed family)
  • •pip install -U transformers==5.18.0; docs for diarization/omni/gte linked above
Read details→
A
Anthropicanthropics·38m ago
🛠️ Tooling

Claude Code 2.1.286: stacked permission counts; resume keeps parallel-tool turns; secret redaction & model-fallback hardening

Anthropic shipped [`Claude Code 2.1.286`](https://github.com/anthropics/claude-code/releases/tag/v2.1.286) on 2026-09-30 (npm [`@anthropic-ai/[email protected]`](https://www.npmjs.com/package/@anthropic-ai/claude-code)): permission prompts show stacked counts like "2 of 5"; `--resume`/`--continue` no longer drop turns after parallel tool-call crashes; broader secret/Bearer/URL-password redaction; one same-tier prior-model retry when the API refuses the default/alias; Remote Control disconnects when org policy disables it; Agent SDK TS [`0.3.286`](https://github.com/anthropics/claude-agent-sdk-typescript/releases/tag/v0.3.286) / Python [`0.2.163`](https://github.com/anthropics/claude-agent-sdk-python/releases/tag/v0.2.163) parity.

Claude Code 2.1.286: stacked permission counts; resume keeps parallel-tool turns; secret redaction & model-fallback hardening
⚡ Key Takeaways
  • •Release v2.1.286; npm ~2026-09-30T17:14:38Z
  • •Stacked permission prompts show counts like "2 of 5"; fullscreen "N more" rows are mouse-clickable
  • •--resume/--continue keep turns after parallel tool-call crashes; non-text tool/hook returns no longer 400
  • •Secret redaction hardened for Bearer/Basic, URL passwords, ZWSP key names; /feedback zip JSON fixed
  • •Agent SDK TS 0.3.286 / Python 0.2.163 parity; docs https://docs.claude.com/en/docs/claude-code/overview
Read details→
A
Anthropicanthropics·39m ago
🚀 Release

Anthropic TS SDK 0.130 / Python 1.10: Admin analytics & plugin marketplaces; Organization API GA; MCP tunnels beta

Anthropic shipped [`@anthropic-ai/[email protected]`](https://github.com/anthropics/anthropic-sdk-typescript/releases/tag/sdk-v0.130.0) and Python [`anthropic==1.10.0`](https://github.com/anthropics/anthropic-sdk-python/releases/tag/v1.10.0) on 2026-09-30: Admin API gains enterprise analytics, spend limits, RBAC groups/roles, Plugins/Plugin Marketplaces, and per-user usage/cost reports; Organization API endpoints go GA; MCP tunnels beta returns read-only `transport` plus one-time relay `token` on create; Managed Agents add refusal stop reason and repository error types.

Anthropic TS SDK 0.130 / Python 1.10: Admin analytics & plugin marketplaces; Organization API GA; MCP tunnels beta
⚡ Key Takeaways
  • •Releases: TS sdk-v0.130.0 at 2026-09-30T18:29:28Z; Python v1.10.0 at 2026-09-30T18:28:45Z
  • •Admin API: analytics, spend limits, RBAC, Plugins/Marketplaces, per-user usage/cost; compare
  • •Organization API endpoints now GA; MCP tunnels beta adds read-only transport + one-time relay token
  • •Managed Agents: refusal stop reason/stop_details on idle; repository error types
  • •Install: npm i @anthropic-ai/[email protected] / pip install anthropic==1.10.0; docs https://docs.anthropic.com/en/api/admin-api
Read details→
L
LangChainlangchain-ai·5h ago
🚀 Release

LangSmith SDK 0.14.2: Python ls.address agent handle, JS OTel tool fields, primary replica keeps original run ids

LangSmith SDK v0.14.2 (2026-09-30): Python ls.address (#3581), JS OTel tool fields (#3592), primary replica keeps original run ids (#3623).

LangSmith SDK 0.14.2: Python ls.address agent handle, JS OTel tool fields, primary replica keeps original run ids
⚡ Key Takeaways
  • •Release v0.14.2 at 2026-09-30T14:41:07Z; PyPI langsmith 0.14.2
  • •Python ls.address agent addressing (#3581)
  • •JS OTel tool name/call_id/definitions (#3592)
  • •Only primary write replica keeps original run ids (#3623)
  • •Compare ahead_by 8; docs https://docs.smith.langchain.com/
Read details→
ADSponsored
L
LangChainlangchain-ai·5h ago
🚀 Release

LangChain OpenAI 1.6.7: AzureChatOpenAI auto-discovers AKS Workload Identity without hand-rolled token providers

LangChain shipped langchain-openai==1.6.7 on 2026-09-30: PR #40532 auto-discovers Azure Workload Identity for AzureChatOpenAI; optional azure-identity extra; explicit credentials still win.

LangChain OpenAI 1.6.7: AzureChatOpenAI auto-discovers AKS Workload Identity without hand-rolled token providers
⚡ Key Takeaways
Read details→
M
Mastramastra-ai·5h ago
🚀 Release

Mastra @mastra/core 1.72.0: live channel resolvers, lease-fenced background tasks, Microsoft Teams channel

Mastra shipped [`@mastra/[email protected]`](https://github.com/mastra-ai/mastra/releases/tag/%40mastra/core%401.72.0) on 2026-09-30: live channel resolvers, lease-fenced background tasks, durable/evented recovery, aggregateTraces(), and @mastra/teams. Breaking: respondToToolApproval requires toolCallId.

Mastra @mastra/core 1.72.0: live channel resolvers, lease-fenced background tasks, Microsoft Teams channel
⚡ Key Takeaways
  • •Release @mastra/[email protected] at 2026-09-30T10:31:27Z
  • •Live channels() resolvers + resolveChannels() avoid redeploys
  • •Lease-fenced background tasks with ownerId + expectedOwnerId writes
  • •New @mastra/[email protected] and 25 Teams tools in Connect
  • •Breaking: respondToToolApproval requires toolCallId
Read details→
B
BerriAIBerriAI·9h ago
🚀 Release

LiteLLM 1.103.1: bind UI/CLI session tokens to their own AES-GCM context (1.103.x security backport)

LiteLLM [`v1.103.1`](https://github.com/BerriAI/litellm/releases/tag/v1.103.1) (PyPI ~2026-09-29T23:36Z; **5** commits ahead of 1.103.0) backports binding UI/CLI session tokens to their own AES-GCM encryption context (`a80eaca`), with e2e/OIDC and auth unit tests—security hardening for operators pinned on the 1.103.x stable line.

LiteLLM 1.103.1: bind UI/CLI session tokens to their own AES-GCM context (1.103.x security backport)
⚡ Key Takeaways
Read details→
A
Anthropicanthropics·9h ago
🚀 Release

Anthropic TypeScript SDK 0.129.0: claude-sonnet-5-5, between_tools thinking, start tool calls while the reply streams

Anthropic shipped [`@anthropic-ai/[email protected]`](https://github.com/anthropics/anthropic-sdk-typescript/releases/tag/sdk-v0.129.0) on 2026-09-28 (**36** commits ahead of 0.128.0): `claude-sonnet-5-5`, `between_tools` thinking, cache diagnostics GA, Managed Agents event type filters, `ClientToolUnion`, and optionally starting tool calls while the reply streams—plus streaming tokenizer/partial tool-input fixes. Mirrors the Python 1.9.0 surface for TS agents.

Anthropic TypeScript SDK 0.129.0: claude-sonnet-5-5, between_tools thinking, start tool calls while the reply streams
⚡ Key Takeaways
Read details→
ADSponsored
V
Vercelvercel·9h ago
🚀 Release

Vercel AI SDK [email protected]: idle SSE heartbeats; prune tool content when retaining zero messages; user messages supersede pending tool approvals

Vercel shipped [`[email protected]`](https://github.com/vercel/ai/releases/tag/[email protected]) on 2026-09-30 ([compare](https://github.com/vercel/ai/compare/[email protected]@7.0.123)): optional SSE heartbeats keep idle UI message streams open; prune all tool content when retaining zero trailing messages; ignore pending tool approvals superseded by user messages; preserve partial reasoning tags at streamed text ends. Build migrates tsup→tsdown with gateway/provider/utils bumps.

Vercel AI SDK ai@7.0.123: idle SSE heartbeats; prune tool content when retaining zero messages; user messages supersede pending tool approvals
⚡ Key Takeaways
Read details→
e
earendil-works@earendil-works·14h ago
🚀 Release

Pi 0.99.1: OpenAI Codex default model switches to GPT-6.1 Sol; bundled /login module fix

Coding-agent toolkit [Pi](https://github.com/earendil-works/pi) ([v0.99.1](https://github.com/earendil-works/pi/releases/tag/v0.99.1), 2026-09-29) adds `gpt-6.1-sol` to OpenAI, Azure OpenAI Responses, and OpenAI Codex providers, makes GPT-6.1 Sol the **default OpenAI Codex model**, and fixes bundled `/login` failing with a missing `openai-chatgpt.js` module. **5** commits ahead of [v0.99.0](https://github.com/earendil-works/pi/compare/v0.99.0...v0.99.1). Model docs: [models.md](https://github.com/earendil-works/pi/blob/v0.99.1/packages/coding-agent/docs/models.md).

Pi 0.99.1: OpenAI Codex default model switches to GPT-6.1 Sol; bundled /login module fix
⚡ Key Takeaways
  • •Release v0.99.1; 5 commits ahead of v0.99.0
  • •OpenAI Codex default → GPT-6.1 Sol (gpt-6.1-sol); also on OpenAI + Azure OpenAI Responses
  • •Bundled /login missing openai-chatgpt.js fixed
  • •Docs: models.md; catalog https://pi.dev/models
  • •Upgrade to v0.99.1; verify with /model; pin via Ctrl+S
Read details→
O
OpenAI@openai·14h ago
🚀 Release

OpenAI Node 7.25.0: Computer Use on beta.agents; 7.24 adds GPT-6.1 Sol + Agents credential/session options

Official TypeScript SDK [`[email protected]`](https://www.npmjs.com/package/openai) ([v7.25.0](https://github.com/openai/openai-node/releases/tag/v7.25.0), 2026-09-29) adds Computer Use to `beta.agents` ([#2837](https://github.com/openai/openai-node/pull/2837)), mirroring Python 3.22.0. Prior [v7.24.0](https://github.com/openai/openai-node/releases/tag/v7.24.0) shipped Agents credential/session options ([#2812](https://github.com/openai/openai-node/pull/2812)), the `gpt-6.1-sol` model id ([#2836](https://github.com/openai/openai-node/pull/2836)), Responses Cyber access, Realtime Translation client secrets, and a dedicated Node translation WebSocket. **35** commits ahead of [v7.23.0](https://github.com/openai/openai-node/compare/v7.23.0...v7.25.0).

OpenAI Node 7.25.0: Computer Use on beta.agents; 7.24 adds GPT-6.1 Sol + Agents credential/session options
⚡ Key Takeaways
  • •Releases v7.24.0 → v7.25.0; npm [email protected]; 35 commits ahead of v7.23.0
  • •Computer Use on beta.agents via #2837
  • •gpt-6.1-sol id (#2836); Agents credential/session options (#2812)
  • •Realtime Translation client secret (#2819) + Node translation WebSocket (#2818)
  • •Upgrade: npm i [email protected]; docs: platform.openai.com agents + API reference
Read details→
C
Cline@cline·14h ago
🚀 Release

Cline 4.1.22: GPT-6.1 Sol becomes default across gateways; Anthropic server-side refusal fallback; Bee & Pareto providers

Cline shipped [v4.1.22](https://github.com/cline/cline/releases/tag/v4.1.22) on 2026-09-30 (**59** commits ahead of v4.1.21): refreshed catalog makes GPT-6.1 Sol the default for OpenAI, OpenRouter, GitHub Copilot, Requesty and more when unpinned; direct Anthropic uses server-side refusal fallback with OpenRouter/Cline failover; adds Bee (HEOSSI) and Pareto Inference (`PARETO_API_KEY`); fixes cancel races, double-counted reasoning tokens, and Bedrock inference-profile routing. SDK [@cline/[email protected]](https://www.npmjs.com/package/@cline/core) ships the same day.

Cline 4.1.22: GPT-6.1 Sol becomes default across gateways; Anthropic server-side refusal fallback; Bee & Pareto providers
⚡ Key Takeaways
  • •Release v4.1.22; 59 commits ahead of v4.1.21
  • •Unpinned defaults: GPT-6.1 Sol on OpenAI/OpenRouter/Copilot/Requesty and more; Vercel AI Gateway→Ling 3.1 Flash; CrossModel/Ofox→Sonnet 5.5
  • •Anthropic server-side refusal fallback + OpenRouter/Cline allow_fallovers; content-filter finish reason instead of empty response
  • •New providers: Bee (HEOSSI), Pareto Inference (PARETO_API_KEY)
  • •SDK @cline/[email protected]; docs https://docs.cline.bot/
Read details→
O
OpenAI@openai·19h ago
🚀 Release

OpenAI Codex 0.159.2: suppress Windows console window flashing for background/sandbox processes

OpenAI Codex CLI shipped rust-v0.159.2 (npm `@openai/[email protected]`): backport [#49385](https://github.com/openai/codex/pull/49385) suppresses Windows console window flashing when launching background processes and sandboxed commands (tracks [#48074](https://github.com/openai/codex/issues/48074)). Diff vs v0.159.1: ~**2 commits / 47 files**. Default model remains GPT-6.1 Sol from 0.159.1.

OpenAI Codex 0.159.2: suppress Windows console window flashing for background/sandbox processes
⚡ Key Takeaways
  • •Release rust-v0.159.2 / npm 0.159.2
  • •Windows: suppress console flashes for background + sandboxed commands (#49385 / #48074)
  • •~2 commits / 47 files vs v0.159.1
  • •Default model still GPT-6.1 Sol from 0.159.1
  • •Upgrade: npm i -g @openai/[email protected]; smoke-test on Windows sandbox hosts
Read details→
V
Vercel@vercel·19h ago
🚀 Release

Vercel AI SDK: `@ai-sdk/openai` 3.0.121 / 2.0.132 add native GPT-6.1 Sol support

Vercel AI SDK shipped `@ai-sdk/[email protected]` and `@ai-sdk/[email protected]` on 2026-09-30, adding native **GPT-6.1 Sol** (`gpt-6.1-sol`) support on the v5/v6 provider lines ([#21747](https://github.com/vercel/ai/pull/21747), backport [#21749](https://github.com/vercel/ai/pull/21749)). Official pricing: **$2/$10** per 1M tokens (cached input $0.10), **1.05M** context, **128k** max output.

Vercel AI SDK: `@ai-sdk/openai` 3.0.121 / 2.0.132 add native GPT-6.1 Sol support
⚡ Key Takeaways
  • •Shipped @ai-sdk/[email protected] and @2.0.132 with GPT-6.1 Sol (#21747 / #21749)
  • •Model id gpt-6.1-sol; specs $2/$10 per 1M, cached input $0.10, 1.05M context, 128k max out
  • •Mid-tier coding/computer-use path vs Astra; aligns with Codex defaulting to Sol
  • •Upgrade: npm i @ai-sdk/[email protected] (or 2.0.132 on v5 line)
  • •Docs: OpenAI provider
Read details→
P
Pydantic@pydantic·19h ago
🚀 Release

Pydantic AI 2.52.0: unified Workspaces for Coder/Shell; web_fetch HTML DoS fix; CLAI2 ships

Pydantic AI shipped v2.52.0 on 2026-09-30 (`pydantic-ai==2.52.0`): `ctx.workspace` unifies local/sandbox file+command APIs with E2B, Fly.io Sprites, Modal, SSH, and Bubblewrap backends; harness moves in-repo to 0.52.0; first `pydantic-clai2` release (`uvx pydantic-clai2`). Also patches local `web_fetch` nested-HTML resource exhaustion (GHSA-v36g-jcw9-x7cw, CVSS 6.5) and adds Claude Sonnet 5.5 plus GPT-6.1 Sol support.

Pydantic AI 2.52.0: unified Workspaces for Coder/Shell; web_fetch HTML DoS fix; CLAI2 ships
⚡ Key Takeaways
  • •Release v2.52.0; 533 commits ahead of v2.51.0
  • •Security: local web_fetch nested-HTML DoS patched (GHSA-v36g-jcw9-x7cw, CVSS 6.5); v1 patch 1.107.7; provider-native fetch unaffected
  • •Workspaces via ctx.workspace (#6492) with E2B/Sprites/Modal/SSH/Bubblewrap backends; harness in-repo at 0.52.0; CLAI2 first release
  • •Models: Claude Sonnet 5.5 (#8974), GPT-6.1 Sol (#9305); Anthropic default max_tokens 16384 on Sonnet 4.5+ (#9025)
  • •Upgrade: pip install -U 'pydantic-ai==2.52.0'; uvx pydantic-clai2; docs: Workspace + Harness
Read details→
Y
Yusuke Wada@yusukebe·22h ago
🛠️ Tooling

Cloudflare launches agentic CLI `cf`: 3000+ API ops from OpenAPI, JSON-default + cloudflare.config.ts

Hono creator @yusukebe amplified Cloudflare’s launch of `cf` (`npm i -g cf`): Forge generates 3000+ CLI ops from OpenAPI (vs Wrangler’s ~280), JSON-default output, `cf cli search`, TypeScript `cloudflare.config.ts`, and Vite-default local dev. Open beta `[email protected]`.

Cloudflare launches agentic CLI `cf`: 3000+ API ops from OpenAPI, JSON-default + cloudflare.config.ts
⚡ Key Takeaways
  • •Blog: Introducing cf (2026-09-28) — agents were 48% of Wrangler use last week; surface expands ~280 → 3000+ ops
  • •Install: npm i -g cf (1.0.0-beta.5); key cmds cf deploy / cf migrate / cf init / cf cli search
  • •Config: TypeScript cloudflare.config.ts with defineConfig / bindings / triggers; Vite-default local DX
  • •Pipeline: Forge (Apache 2.0, cloudflare/forge); CLI at cloudflare/cf
  • •Migration: final Wrangler major will point to cf, then 18 months maintenance; no model SWE-Bench numbers for this CLI
Read details→
O
OpenAI@OpenAI·22h ago
🛠️ Tooling

Personal desktop agents compared: Grok Bot vs Meta Muse vs OpenAI Dots

After OpenAI launched Dots (GPT-6 Astra, per-dot cloud computer, 4000+ plugins) at DevDay, the personal/desktop agent space has three clear poles: Cursor/xAI Grok Bot (developer teammates + routines), Meta Muse (Secure VM consumer agent; often misheard as “Mate Muse”), and OpenAI Dots (always-on primary Dot). Feature comparison from official docs only—no invented benchmarks.

Personal desktop agents compared: Grok Bot vs Meta Muse vs OpenAI Dots
⚡ Key Takeaways
  • •“Mate Muse” verifies as Meta Muse (2026-09-08): Meta news + How We Designed Muse; Muse Secure VM + Muse Spark; consumer life/goals agent.
  • •OpenAI Dots (DevDay 2026-09-29): Introducing dots; GPT-6 Astra; own cloud computer; 4000+ plugins; ChatGPT Pro / Business Premium; one primary Dot at launch.
  • •Grok Bot: Cursor docs — persistent cloud computer, plugins/MCP, skills/routines, multi-Bot; included with paid Cursor plans or SuperGrok / X Premium+ link (Plans).
  • •No public three-way official SWE-Bench/Agent Arena bake-off; pick by job: coding/routines/multi-Bot → Grok Bot; consumer life agent → Muse; ChatGPT ecosystem always-on → Dots.
  • •Shared pattern: isolated cloud computer + approval for sensitive acts; differ on developer routines/local exec (Grok Bot), WhatsApp/consumer connectors (Muse), plugin breadth + Pro bundling (Dots).
Read details→
O
OpenAI@OpenAI·22h ago
🚀 Release

OpenAI launches Dots: always-on GPT-6 Astra agents with a cloud computer across 4,000+ apps

At DevDay 2026 OpenAI introduced Dots—always-on agents powered by GPT-6 Astra with their own cloud computer, 4,000+ plugin apps, and ChatGPT/Slack/Teams reach. Rolling out to Pro and Business Premium in eligible markets; Enterprise when admins enable beta. First-month usage doesn't count against plan allowances.

OpenAI launches Dots: always-on GPT-6 Astra agents with a cloud computer across 4,000+ apps
⚡ Key Takeaways
  • •Form factor: always-on agent with its own cloud computer/browser; 4,000+ apps via plugins — Introducing dots
  • •Model: GPT-6 Astra; Auto-review respect eval: Astra 0 bypass attempts vs GPT-5.6 Sol 5% (0.3% weak-config exploit) — Path to Astra
  • •Access: Pro (excl. EEA/Switzerland/UK) + Business Premium rolling out; Enterprise admin-enabled; 1 primary dot at launch; first month excluded from allowances — Help
  • •Controls: Custom Rules + Auto-review; proactive research is read-only; specialist dots target Microsoft Agent 365
  • •Start: create on desktop app/web → name + plugins → optional local computer + Codex cloud tasks; see safety FAQ
Read details→
O
OpenAIopenai·1d ago
🚀 Release

OpenAI Python 3.22.0: Computer Use on beta.agents; browser auth/screenshots + desktop environment templates

OpenAI’s Python SDK shipped v3.22.0 on 2026-09-29 (PyPI `openai==3.22.0`): typed computer-use tools on `beta.agents`, optional screenshots, browser authentication and origin-access approval inputs, plus desktop settings / environment templates for hosted environments. Node SDK [v7.25.0](https://github.com/openai/openai-node/releases/tag/v7.25.0) landed the same API surface the same day.

OpenAI Python 3.22.0: Computer Use on beta.agents; browser auth/screenshots + desktop environment templates
⚡ Key Takeaways
  • •Release v3.22.0; PR #3989
  • •beta.agents gains computer_use tools, screenshots, browser auth + origin-access approvals
  • •Desktop settings / environment templates for hosted agent environments
  • •Node twin: v7.25.0
  • •Upgrade: pip install -U openai==3.22.0; guide: Agents
Read details→
A
Anthropicanthropics·1d ago
🛠️ Tooling

Claude Code 2.1.285: `claude --desktop`; plugin configure & MCP install-time config; allowedProviders lock-down

Anthropic Claude Code shipped v2.1.285 on 2026-09-29 (npm `@anthropic-ai/[email protected]`): `claude --desktop` opens the desktop app on the current directory or a resumed session; `claude plugin configure` plus `--config server.key=value` set MCP options at install time; managed `allowedProviders` limits which API providers a machine may use; `CLAUDE_CODE_DISABLE_WEB_FETCH` disables WebFetch. ~4 commits / 67 files vs v2.1.284, plus many SSH/sandbox/subagent permission fixes.

Claude Code 2.1.285: `claude --desktop`; plugin configure & MCP install-time config; allowedProviders lock-down
⚡ Key Takeaways
  • •Release v2.1.285; ~4 commits / 67 files vs v2.1.284
  • •claude --desktop opens Claude Desktop on cwd or --continue/--resume
  • •plugin configure + install-time --config server.key=value for bundled MCP settings
  • •Managed allowedProviders + CLAUDE_CODE_DISABLE_WEB_FETCH
  • •Upgrade: npm i -g @anthropic-ai/[email protected]; docs at code.claude.com/docs
Read details→