Ce que Render a publié en août 2026

New compute plans for memory-intensive applications

Render introduced new compute plans tailored for memory-intensive applications, likely expanding their pricing tiers to better serve workloads with high memory demands.

Render introduces memory-optimized compute plans and spec-based IDs

Render launched new compute plans with higher RAM-to-CPU ratios and a 12-CPU tier for web services, private services, and background workers. Existing plans now use spec-based IDs (e.g., `4c-32g`) whi…

Render expands OpenTelemetry metrics with CPU, memory, network, HTTP, disk, and Postgres coverage

Render introduced a comprehensive set of OpenTelemetry metrics for observability, covering CPU, memory, network, HTTP requests, disk usage, and Render Postgres/KV stores. All metrics use the `render.`…

How Render Workflows powers our internal data pipeline

Render describes how their internal data pipeline leverages Render Workflows to streamline processes and improve efficiency. The blog post details technical implementation and benefits.

Render updates language support and versioning controls for native runtimes

Render expanded its documentation to clarify language support for native runtimes, including default versions and methods to pin versions for Node.js, Bun, Python, Ruby, Go, Rust, and Elixir. Go and R…

Render enables deployment of Shopify apps via Docker and CLI workflow

Render now supports deploying custom Shopify apps using its platform, enabling developers to extend Shopify functionality via Render-hosted services. The integration requires setting environment varia…

Render adds OIDC integrations with AWS, Anthropic, and OpenAI for managed auth

Render introduced managed OpenID Connect (OIDC) integrations for AWS, Anthropic, and OpenAI, enabling Pro workspace users to authenticate services automatically with short-lived tokens. Each provider …

Render CLI expands with non-interactive mode and AI agent skills

Render released a major update to its CLI, adding non-interactive mode for CI/CD and automation, expanded service management commands, and support for AI coding tools via agent skills. The CLI now val…

Render updates Bun version defaults and configuration methods

Render changed the default Bun version for new services to 1.3.14 (from 1.3.4) and introduced explicit version control via BUN_VERSION environment variables or .bun-version files. Services created bef…

Pricing change detected for Render

Pricing updated for Render: - Starter: annual price $7/mo → $5/mo - Starter: allowance changed from 1 service to 3 services - Pro: annual price $19/mo → $15/mo - Pro: allowance changed from 1 service …

Four minutes after midnight, Codex said the page was live

Render's blog announces a new feature where their AI assistant, Codex, can deploy a page live within minutes of code completion, reducing deployment time significantly.

Render updates outbound IP routing: services must handle TCP resets

Render now proactively resets TCP connections to public destinations when outbound IP routing changes occur due to maintenance, IP rotation, or dedicated IP adjustments. Services must implement retry …

Stripe: An orchestration layer, not a platform

Render's blog post argues that Stripe functions as an orchestration layer rather than a full platform, positioning itself as a more comprehensive alternative.

Render details secure Postgres design for LLM audit compliance

Render published a technical guide on building an append-only Postgres audit table for LLM prompts and outputs, addressing immutability, retention, access control, and GDPR/HIPAA compliance. The desig…

Render reduces median service build time by 40% across all runtimes

Render cut median service build times by 40% globally by upgrading to faster CPU and disk nodes, reducing the median from 38 seconds to 21 seconds. The improvement applies to all runtimes and follows …

Render publishes Rails migration guide from Railway

Render published a step-by-step guide to migrate Rails applications from Railway to Render, covering service mapping, Blueprint configuration, environment variables, and database migration procedures.…

How Locunity covers democracy, not infrastructure

Render features Locunity as a customer success story highlighting their use of Render's platform for democratic coverage, with an embedded offer for cloud migration credits.

Pricing updated for Render: - Free: allowance changed from 1 service, 512 MB RAM, 1 GB storage to 1 service - Starter: annual price $5/mo → $7/mo - Starter: allowance changed from 1 service, 1 GB RAM,

AssemblyAI: A voice agent can fail without throwing an error

A blog post by AssemblyAI discusses how their voice agent can fail silently without throwing an error, potentially impacting reliability.