Releasesby GlyphGeneral Availability of OCR 4.1
OCR 4.1 (mistral-ocr-4-1) is now Generally Available.
Monday, August 31, 202677 releases
Google Cloud6 products
Gemini 3.1 Flash Image and Gemini 3 Pro Image updates
This release introduces feature updates and expanded endpoint availability for Gemini 3.1 Flash Image and Gemini 3 Pro Image: Multi-region endpoint support : Gemini 3.1 Flash Image ( gemini-3.1-flash-image ) is now available on US ( us ) and EU ( eu ) multi-region endpoints for model availability, ML processing, Provisioned Throughput, and PayGo Standard.
OpenTelemetry support for AI monitoring is now GA
No New Relic agent required. If your app already emits OpenTelemetry traces, you can now send GenAI spans directly to AI monitoring and see every LLM call, tool call, and agent step , as part of your existing distributed traces.
Investigate user sessions end-to-end with the new Session Replay
Dynatrace has generally released a new Session Replay built on the Grail data lakehouse, allowing SaaS customers to replay user experiences alongside logs, traces, and metrics.
Administration
- Added support for customizing the main navigation in Omni. Skills can now be enabled and disabled at the organization level . Omni now supports Canada Central, UAE North, and Southeast Asia cloud regions for Azure deployments.
Custom SMTs can make external network calls (Limited Availability)
Custom SMTs on fully managed connectors can call endpoints outside Confluent Cloud, such as KMS, secret stores, or data enrichment APIs, as a Limited Availability feature.
Selected GitHub Copilot models deprecated
As of today, September 1, 2026, we have deprecated the following models across most GitHub Copilot experiences (including Copilot Chat, inline edits, ask and agent modes, and code completions). Note that Claude Sonnet 4.6 is still available to individual GitHub Copilot subscribers on annual plans.
End of life: Pinecone Marketplace
Pinecone Marketplace has reached end of life and is no longer available. Marketplace ran on Pinecone Assistant. To keep querying your data, see Pinecone Assistant, and contact Support if you need help.
ABAC GRANT policies are generally available for models and services
ABAC GRANT policies for models, model services, model provider services, Model Context Protocol (MCP) services, and agent services are now generally available. These policies dynamically grant Unity Catalog privileges to securable objects whose governed tags match a condition, instead of granting privileges on each object individually.
Pipeline events system table is in Beta
The pipeline_events system table is in Beta . This table records Lakeflow pipelines event log entries for pipelines in your account, capturing lifecycle transitions, flow progress, data quality metrics, errors, and other operational data across all pipelines and workspaces in a region.
Databricks SQL version 2026.36 is now available in Preview
Databricks SQL version 2026.36 is now available in the Preview channel. See features in 2026.32 .
Second generation Openflow (Preview)
Second generation (gen 2) Openflow is now in public preview. Gen 2 makes Openflow deployments, runtimes, and connectors first-class Snowflake SQL objects with standard RBAC, SQL lifecycle management, and versioned connector configuration. Supported connectors at public preview: PostgreSQL CDC, MySQL/MariaDB CDC. Gen 2 Openflow is available on AWS and Azure.
Snowflake-provided tags (Public Preview)
With this preview release, Snowflake-provided tags are available in every account. These out of the box tags in the SNOWFLAKE.TAGS schema give you a consistent vocabulary for common governance use cases, including cost center, certification status, sensitivity, environment, and project, without creating and maintaining equivalent tags in each account.
app.yml for Snowflake App Runtime (Preview)
Snowflake App Runtime projects use a single app.yml file to describe both deployment and the remote build. Include version: 2 at the top of the file. snow app setup writes this file for new projects.
Training New Serverless Training models: DeepSeek V4 Flash 0731, Qwen 3.8 27B, and Muse Glimmer 30B The...
Training New Serverless Training models: DeepSeek V4 Flash 0731, Qwen 3.8 27B, and Muse Glimmer 30B The following models are now available for LoRA workloads on the shared Serverless Training pool: DeepSeek V4 Flash 0731 with up to 262K context Qwen 3.8 27B with up to 128K context Muse Glimmer 30B with up to 128K context See the Serverless Training guide...
Week 36
Automatic data retention telli now supports automatic data retention. If data retention is included in your plan, set how long contacts and call data are kept under Settings > Organization . Choose from 1 to 730 days for contacts, all call data, recordings, or transcripts. telli applies the rules daily and asks for confirmation before saving.
A new Gradium Text-to-Speech model is live and is now the default
A new Gradium Text-to-Speech model is live and is now the default. Existing voices, including custom voices, work as-is. The model reads the structured entities that break voice agents in production, such as phone numbers, email addresses, IBANs and reference codes, with no pre-processing or text normalization needed.
August 24–30, 2026
- Added: You can now request plain HTTP access to the Object Storage S3 API for selected buckets and use http:// instead of https:// in the S3 API endpoint. Choose between the Region-local mode (access only from Nebius AI Cloud VMs in the bucket's region) and the All mode (access from any source).
Wildcard Transform Rules & Server-Side API Key Validation
Wildcard Transform Rules & Server-Side API Key Validation This week improves sandbox networking and authentication with wildcard domains in transform rules, server-side API key validation, and readable proxy errors for browser requests.
The Polaris CDN is adopting semantic versioning
Polaris Web Components loaded from the Shopify CDN are moving to semantic versioning. Most apps will continue receiving compatible improvements automatically, while developers can choose when to adopt breaking changes. Existing apps don’t need to take immediate action.
Cross-App Access Requesting App now in Early Access
We're excited to announce that Auth0 now supports acting as a __requesting application in Cross-App Access flows in Early Access__. Today, every user has to click "Connect your account" for every integration, in every app, one at a time.
Snyk Projects now stay in sync with your repository content
The "What" Repo Content Sync keeps your Snyk Projects aligned with what your repositories actually contain. When a change is merged to a branch Snyk monitors, Snyk creates and begins monitoring Projects for newly added manifest, Dockerfile, and configuration files, and deactivates the Projects whose files were removed.
Browser Run - Crawl endpoint now respects the Content Signals `use` directive
The /crawl endpoint now respects the use directive of the Content Signals ↗ standard, letting site owners express the maximum level at which their content may be used. You can declare your intended level with the new contentUse parameter. Allowed values, from least to most permissive, are reference and full , and the default is full .
Wider speed range for Flux TTS
Flux TTS speed now runs from 0.5 to 1.5 in 0.05 increments, up from 0.85 – 1.15 . The default stays 1.0 , and every previously accepted value still works. The wider range applies everywhere speed is accepted on Flux TTS: the /v2/speak streaming WebSocket (at connection and mid-session with Configure ), the /v2/speak batch REST endpoint, and...
New serverless models
The following models are now available on serverless : Qwen/Qwen3.8-Flash : 1,000,000 context length. Pricing: $0.15 input / $0.47 output (per 1M tokens).
Greenhouse connector moves to OAuth and latest Harvest API
The Greenhouse connector signs in with OAuth and uses Greenhouse's latest Harvest API, adding interviews and file attachments.
ClickPipes: Private Service Connect supports Global Access
ClickPipes now supports Global Access for cross-region private connections using Private Service Connect (PSC). When enabled, ClickPipes can connect to a PSC endpoint in a different region than the GCP region where the service is running, including regions not natively supported in ClickHouse Cloud (*e.g.* asia-northeast3 (Seoul)).
Set per-user budgets on AI Gateway
You can now use AI Gateway budgets to set a dollar spending limit for each user on your team. The limit covers spend from every API key attributed to the user, along with their app tokens. After it's reached, AI Gateway rejects new requests until the budget resets or is increased.
Troubleshoot and secure your code faster with Datadog’s Bitbucket Cloud Source Code integration
Connect Bitbucket Cloud to Datadog to troubleshoot with source code in context and surface test, quality, and security feedback in pull requests.
LanceDB v0.38.0
🛠 Breaking Changes fix: make table existence manifest-authoritative refactor(python): require pydantic v2 fix(nodejs)!: key parsed embedding configs by vector column feat!: rename branch merge to cherry_pick feat(nodejs): materialized view bindings fix: package ordinary @udf bodies and emit only the V1 type grammar fix(node): preserve embedding...
Docling 2.124.0
Feature cli: Add latex output format ( #4101 ) ( c4b0fc7 ) Add terminal task callback event ( #4095 ) ( d2283ad ) Fix service_client: Handle Retry-After HTTP-date with unknown timezone ( #3786 ) ( 3bba8da ) Silence transformers generation-config deprecation warning and add max_new_tokens CLI overrides ( #4115 ) ( 01fc722 ) Drop the UTF-8 BOM when decoding...
Sunday, August 30, 20263 releases
Cortex Agents asynchronous API (General availability)
The Cortex Agents asynchronous API is now generally available on AWS and Microsoft Azure.
AI Search now supports GLM-5.3 Flash
AI Search now supports @cf/zai-org/glm-5.3-flash for text generation. The model has a 1,048,576-token context window and runs on Workers AI. To configure the model for an AI Search instance, refer to Supported models .
MiniMax H3 and H3 Max are 50% off on AI Gateway
MiniMax H3 and H3 Max are 50% off on AI Gateway from August 30 through September 13, in partnership with MiniMax. The discount covers requests billed through AI Gateway, at every duration and in every aspect ratio the model supports.
Saturday, August 29, 20267 releases
OpenAI API mTLS and X.509 workload identity federation are GA
Mutual TLS and X.509 workload identity federation can be configured in the Platform console, with access governed by organization roles and permissions, enabling keyless authentication.
ai_transcribe function (Beta)
The ai_transcribe() SQL function transcribes an audio file into text. It returns the transcript as time-stamped segments, and labels each segment with a speaker when speaker diarization is enabled for your workspace. The output is a VARIANT value, so you can chain it into other AI Functions. This feature is in Beta . See ai_transcribe function .
Edit Image Quality Update
Hey everyone, we've updated our V8.2 edit model for better image quality. If you've had any issues in the last 24 hours give it a new shot and keep giving us feedback. Thanks! More updates soon.
Recommended migrations
- MiniMax M2.7 , migrate to MiniMax M3 GPT OSS 20B , migrate to GPT OSS 120B or Qwen3 8B for lower-latency workloads Kimi K2.6 Turbo / Fast , migrate to Kimi K2.6 (standard mode) Kimi K2.7 Code Fast , migrate to Kimi K2.7 Code (standard mode) DeepSeek V4 Pro , migrate to DeepSeek V4 Pro (0813)
active_sessions autoscaling metric for TTS
Dedicated endpoint autoscaling now accepts active_sessions , which targets concurrent open client WebSocket sessions per replica on a TTS deployment ( AVERAGE_VALUE only). The default inflight_requests metric also covers TTS handlers, so HTTP TTS traffic can keep scaling without a policy change. See Configure autoscaling .
Friday, August 28, 202642 releases
Training Serverless Training deprecation: Qwen 3.5 9B and Qwen 3.6 27B Qwen 3.5 9B and Qwen 3.6 27B are...
Training Serverless Training deprecation: Qwen 3.5 9B and Qwen 3.6 27B Qwen 3.5 9B and Qwen 3.6 27B are deprecated from Serverless Training effective August 26, 2026. Migrate new and existing Serverless Training workloads to Qwen 3.8 27B . This change applies to the shared Serverless Training pool.
CLI and SDK rollup: pulumi new, pulumi do, and more
We shipped a bunch of new features and improvements in the Pulumi CLI and SDK this month. A few highlights: A redesigned interactive pulumi new that asks for a cloud provider and language instead of listing every template, then confirms your project, stack, and config defaults in a single step ( v3.258 , v3.259 ) Opt-in encryption of stored credentials with...
Dynamic Highlights (research preview)
Dynamic Highlights (research preview) Dynamic Highlights selects excerpts across the complete result set instead of treating each page independently. It gives more of the shared context budget to useful sources and gives less context to sources that only repeat information already returned.
Custom Token Exchange is now Generally Available
We are excited to announce that Custom Token Exchange is now Generally Available. This feature is available to all Enterprise, B2B Professional, and B2C Professional customers, as well as part of the Trial period for free tenants.
What's new in Sigma
AI Allow Assistant to search all sources (Beta) An admin can configure the default behavior of Sigma Assistant to use any data sources available to users or limit Assistant to only specific configured data sources. Users can always manually choose a source for Assistant to use from the sources that they have access to.
Upcoming changes to GitHub Copilot policies and billing
To provide a strong, consistent Copilot experience, we’re making three separate, upcoming changes to Copilot policies and billing. Please review the upcoming updates to understand what may impact you.
Hy4 Preview now available on AI Gateway
Hy4 Preview from Tencent is now available on AI Gateway. Hy4 Preview is an open-source Mixture-of-Experts model with 770B total parameters and 49B active per token, aimed at long-horizon coding, document analysis, game development, and scientific reasoning. It serves a context window of 1M tokens.
Vercel CLI expands commands for DNS, domains, and projects
The Vercel CLI now provides dedicated commands for managing DNS records, domains, and projects. This brings more of the functionality available through the Vercel dashboard and API to the terminal, where it can be used interactively, in scripts, or by agents.
Looker Google Maps Enhancements preview adds dual-axis points and TopoJSON layers
The Dual-axis Map option now supports points and circles, and custom map layers can be loaded from a TopoJSON file URL.
Gemini Enterprise: New data stores and support for new actions (Preview)
The following data stores are available in Public Preview in Gemini Enterprise: Campfire Clay CourtListener Daloopa MSCI You can search and read data from these data stores using natural language. Additionally, the following data stores support new actions in Public Preview: AirOps : Update knowledge base document metadata.
Spanner supports expression indexes on scalar expressions
Secondary indexes can be built on scalar expressions over one or more columns, removing the need for dedicated generated columns just for indexing.
Unity Catalog Skills (Beta)
You can now publish, govern, and share agent skills with Unity Catalog. A skill is a first-class Unity Catalog securable ( catalog.schema.skill ) that follows the SKILL.md specification , so you register it once and control who can discover, use, and update it with the same permissions, tags, and audit that protect your tables.
Monitor and cap external provider costs
Unity Gateway can now track and cap spend on external model providers, such as Amazon Bedrock and Azure AI Foundry, not just Databricks-provided models. If your organization already has capacity commitments directly with a provider, Unity Gateway becomes the single point to track that usage and enforce spend limits across teams and use cases.
Cloudflare One Client for Linux (version 2026.7.1377.0)
A new GA release for the Linux Cloudflare One Client is now available on the stable releases downloads page . This hotfix resolves an issue where a small but noticeable percentage of DNS queries fail across platforms.
Xcode 27.0 Beta 6 Available
The Xcode 27.0 Beta 6 image is now available to be used in CircleCI pipelines. The full manifest of installed software can be found here .
Week 35
Annotate call transcripts You can now annotate a specific moment while reviewing a call transcript in Conversations . Hover an agent, contact, or tool turn to report an issue inline, then adjust the highlighted audio range and describe what went wrong.
More resilient refreshes for expiring offline access tokens
Expiring offline access token refreshes now retain the previously used refresh token until an app begins using its replacement. This rollout helps apps recover when a refresh response is lost or isn’t persisted. No migration is required, but apps should continue to store every refreshed token pair atomically.
Share Email API
Create a shareable link to view a sent or received email directly from the API.
August 2026 by company 127
Google Cloud162
Databricks138
Cloudflare104
Vercel104
Cognition93
GitHub71
Snowflake54
Omni42
Microsoft41
ElevenLabs40
Airbyte39
Deepgram38
Anthropic31
OpenAI31
Grafana Labs28
CrewAI23
Arize AI22
DigitalOcean22
Confluent20
dbt Labs19
Retool18
Feldera16
CircleCI15
Palantir15
PlanetScale14
Tailscale14
Together AI14
ClickHouse12
Hightouch12
Resend12
Snyk12
Black Forest Labs11
Fivetran11
Netlify11
Twilio11
Dynatrace10
Figma10
MongoDB10
MotherDuck10
Datadog9
Hugging Face9
Oracle9
Amplitude8
Okta8
Sentry8
Zed8
Sourcegraph7
Supabase7
Astral6
Oxylabs6
Pinecone6
Pulumi6
Redpanda6
Astronomer5
Cursor5
E2B5
Fireworks AI5
Gradium5
Hex5
Mistral AI5
Nebius5
Prime Intellect5
telli5
Apple4
CoreWeave4
DuckDB4
IBM4
Mixpanel4
Neon4
Notion4
OpenRouter4
Prefect4
Railway4
Replit4
Sigma Computing4
Starburst4
Unstructured4
Windsurf4
Anyscale3
AWS3
Baseten3
Citus Data3
Docker3
Elastic3
fal3
Hasura3
Keenable3
Mem03
Midjourney3
New Relic3
Promptfoo3
Aqua Security2
Expo2
LanceDB2
Linear2
Meta2
Modal2
Modular2
Perplexity2
Render2
Antfly1
Bun1
Cerebras1
Cockroach Labs1
Cohere1
Google DeepMind1
dltHub1
Electric1
Estuary1
Exa1
Fastly1
GitLab1
Guardrails AI1
HashiCorp1
LangChain1
LlamaIndex1
NVIDIA1
PagerDuty1
Percona1
PingCAP1
Posit1
Qdrant1
Redis1
Replicate1
Searchable1
Voyage AI1