AI.info
AI Tools Updates — Page 66
What changed in the AI tools this week — dated, from the makers’ own changelogs.
Auf Englisch
- Devin — October 14, 2026
- Truss — Muse Spark Open and Nemotron 3.5 Super VL available on Baseten
Muse Spark Open and Nemotron 3.5 Super VL are now available through Baseten Model APIs.
- incident.io On-call — The incident.io plugin: our skills, in your agent
A plugin lets coding agents use the powers of incident.io agents.
- incident.io Incident Response — The incident.io plugin: our skills, in your agent
Launched an incident.io plugin for coding harnesses, giving agents the same powers as incident.io's agents.
- Claude Code — 2.1.301
Added a retry usage-limit wait cap, vim bracket-jump motion, multiline mod input, and OpenTelemetry tracing and event fields.
- Pulumi IaC — Pulumi ESC now issues OIDC tokens for any service
Pulumi ESC can now issue short-lived OIDC tokens for any service that accepts them using fn::open::oidc.
- Quorum — Introducing Quincy Hub: The Command Center for AI Agents in Government Affairs
Introducing Quincy Hub, the command center for AI agents in government affairs.
- Avoma Conversation & Revenue Intelligence — Avoma Insider: October 2026 Product Updates
October updates include scheduled AI reports, pre-meeting briefs, mobile tools, Salesforce field mapping, and scorecard and forecast controls.
- browser-use — 0.13.12: OpenAI browser tools
Added OpenAI browser tools backed by Browser Use Actor, supporting local browsers, Browser Use Cloud, and connected BrowserSessions.
- browser-use — 0.13.13: OpenAI browser tools and Cloud fixes
Added an OpenAI browser-tool driver with 18 default tools and six optional actions; fixed Cloud tab URL and scaled-display alignment issues.
- Hermes Agent — Hermes Agent v0.22.2
Fixed supervised gateway startup after a reboot.
- Devin Desktop — October 14, 2026
- Intercom Fin — Search inside a Messenger conversation
Search conversation from the conversation menu to find text in the chat you’re viewing.
- Intercom Fin — Workforce Management
Forecast volume, build schedules and see where you're short-staffed, all inside Intercom.
- Atlan — Empower your assets with clickable buttons- with the new URL badges
URL custom metadata badges now appear as clickable buttons linking to the asset's destination.
- Composio — py@0.28.0
Released Python SDK 0.28.0.
- Dia — Simpler skills, sharper windows
Skills come to Dia for Windows; title-bar buttons can move windows; connected-app tools refresh without an app update.
- Dia — Easier connections, quicker answers
Imports GitHub Live Folders as Dia Live Groups; Chat reads the current page right away; answers simple questions directly.
- Palantir Foundry — Claude Haiku 5.5 now available in AIP
Anthropic’s Claude Haiku 5.5 is now available in AIP for eligible commercial enrollments and US government enrollments at IL2 and IL4.
- Palantir Foundry — Gemini 4 Argon from Google Vertex AI now available in AIP
Gemini 4 Argon from Google Vertex AI is now available as an experimental model in AIP.
- Palantir Foundry — Developer Console now supports Global Branching
Develop, test, and review Ontology SDK (OSDK) React application changes alongside Ontology changes on the same global branch.
- Palantir Foundry — GLM-5.3 now available in AIP through Amazon Bedrock
Z.ai’s GLM-5.3 is available as an experimental model in AIP through Amazon Bedrock for non-georestricted and US-georestricted commercial enrollments.
- Palantir Foundry — GPT-6.1 Sol now available in AIP
OpenAI’s GPT-6.1 Sol is now available in AIP through Direct OpenAI, Azure OpenAI, and Amazon Bedrock for eligible commercial enrollments.
- Palantir Foundry — View state break causes directly in Pipeline Builder
Pipeline Builder's deploy menu now shows which changes caused a state break for streaming and incremental pipelines.
- Gram — See which directory groups grant a role, and add MCP servers on your own network
Role pages show which directory groups and attribute values grant roles; private-network MCP servers can be added.
- Gram — One MCP URL now connects every server in your organization
A single organization-wide MCP URL provides access to reachable servers across projects, and admins can withhold it without disabling sign-in.
- Gram — Signing in to an existing organization now starts your platform trial
Signing in from another Speakeasy product starts a platform trial; setup errors explain themselves and permissions are rechecked live.
- Mistral AI — Introducing Timestral
Introduced Timestral, a probabilistic forecasting model in four sizes (5M–209M parameters).
- Pinecone — On-demand indexes in BYOC
BYOC now supports on-demand indexes in addition to dedicated read nodes indexes.
- Lunit SCOPE IO — Lunit SCOPE Reveals HER2 Expression Differences Linked to Treatment Response Beyond Conventional Testing in Nature Medicine Study
Reveals HER2 expression differences linked to treatment response beyond conventional testing in Nature Medicine study.
- Snowflake Data Cloud — Data mirroring for Snowflake Postgres (General availability)
Data mirroring for Snowflake Postgres
- Flux — FLUX 3 Image Klein
A 5.6B parameter open weights model for text-to-image, multi-reference generation and image editing.
- beehiiv — Connect Your Whole Stack
Connect beehiiv to the tools you use to grow, sell, and run your business.
- Unsloth — NVIDIA Nemotron 3.5 + Meta Muse Spark Open
Added local Nemotron 3.5 and Muse Spark Open models, text-diffusion and DeepSpeed training, and coding-agent prompt caching.
- Unsloth — Flash-Attention2, Causal-Conv1D, Mamba_SSM Binaries
Released prebuilt CUDA 13 wheels for flash-attn, causal-conv1d, and mamba-ssm on Linux and Windows.
- Baseten — Muse Spark Open and Nemotron 3.5 Super VL available on Baseten
Muse Spark Open and Nemotron 3.5 Super VL are now available through Baseten Model APIs.
- TT-Metalium — v0.81.0-dev20261015
Avoids throttled matmul MOP on non-full tiles, bounds fp32-to-uint8 typecast shift, and propagates NaN through the fp32 GELU derivative.
- Ada AI Agent — Opening line for Voice Proactive Outreach
A Voice Proactive Outreach that runs a Playbook can now start every call with an Opening line before the Playbook runs.
- Ada AI Agent — Handoff reasons in the Data Export API and MCP
The v2 Conversations endpoint returns a handoffs array; get_conversation returns reason_code and reasoning.text on handoff entries.
- Kilo Code — v7.8.12 (release)
Routes gateway traffic through ai-gateway.kilo.ai by default, adds a URL override, and changes ignored-path handling in glob and grep.