Releasesby GlyphValidation Errors Now Return 400 Instead of 422 API requests that fail validation now return HTTP 400 Bad...
Validation Errors Now Return 400 Instead of 422 API requests that fail validation now return HTTP 400 Bad Request instead of 422 Unprocessable Entity, aligning the HTTP status code with standard behavior across major inference providers.
Also shipped Apr 24, 2026Cerebras in April 2026 →
Agent SDK: Building Multi-turn Agent Workflows on OpenRouter
The OpenRouter Agent SDK provides a callModel function to build multi-turn agent workflows. It supports tool calls, stop conditions, and cost tracking across over 400 models.
Build Your Own Harness with the Agent SDK
The OpenRouter Agent SDK allows developers to scaffold personalized coding agents with terminal UI or headless support for scripts and pipelines.
What's new in Sigma
Admin AWS signature V4 authentication for API connectors API connectors now support the AWS signature V4 authentication method. For more information, see Configure API credentials and connectors in Sigma .
RLS Tester in Feature Preview
The Supabase Dashboard has added an RLS Tester in Feature Preview, allowing users to run SELECT queries as any role, see which RLS policies fire, and debug policies.
Unity Catalog OSS Unity Catalog AI 0.4.0
The 0.4.0 release of Unity Catalog AI includes several improvements, including expanded Databricks client options and dependency/security improvements. Artifacts in this release unitycatalog-ai unitycatalog-langchain unitycatalog-openai Major Changes Warehouse ID Support in DatabricksFunctionClient : For workspaces where serverless compute is unavailable or...
Organization spend limits, snapshot storage consumption API, One-click Neon MCP setup for Kiro, durable AI agents, and agent-readable docs release - Apr 24, 2026
Organization spend limits and email alerts. You can now set a monthly spending limit for your organization from the Billing page in the Neon Console. When spend...
More from Cerebras InferenceEvery Cerebras Inference release →
New parameter: prompt_cache_key /v1/chat/completions and /v1/completions now accept an optional...
New parameter: prompt_cache_key /v1/chat/completions and /v1/completions now accept an optional prompt_cache_key parameter. Requests that share the same key are routed together so they reuse the same prompt cache, increasing cache hits and reducing time to first token.
New dedicated models: GLM 5, GLM 5.1, and Kimi K2.6 Z.AI's GLM 5 and GLM 5.1 and Moonshot AI's Kimi K2.6 are...
New dedicated models: GLM 5, GLM 5.1, and Kimi K2.6 Z.AI's GLM 5 and GLM 5.1 and Moonshot AI's Kimi K2.6 are now supported on Dedicated Endpoints.
Projects is now generally available Projects is out of private preview and available to all organizations
Projects is now generally available Projects is out of private preview and available to all organizations. Use Projects to group API keys, set per-project rate limits, segment usage analytics, and manage member access across isolated workspaces. Every organization starts with a Default Project, so existing setups are unchanged.
Payload Optimization: msgpack and gzip /v1/chat/completions and /v1/completions now accept request bodies...
Payload Optimization: msgpack and gzip /v1/chat/completions and /v1/completions now accept request bodies encoded with msgpack (Content-Type: application/vnd.msgpack) and/or compressed with gzip (Content-Encoding: gzip). This can significantly reduce payload size and improve TTFT for requests with long prompts.
New Sampling Parameters The following parameters are now available on all models: - frequency_penalty –...
New Sampling Parameters The following parameters are now available on all models: frequency_penalty – Reduces repetition by penalizing tokens based on how often they have appeared so far presence_penalty – Reduces repetition by penalizing tokens that have appeared at least once logit_bias – Adjusts the likelihood of specific tokens appearing in the response
Streaming: Server-Sent Events (SSE) Streaming responses for Chat Completions (/v1/chat/completions) and...
Streaming: Server-Sent Events (SSE) Streaming responses for Chat Completions (/v1/chat/completions) and Completions (/v1/completions) now emit a standard data: [DONE] SSE event at the end of the stream, following the SSE standard (see the living specification). The [DONE] marker is guaranteed to appear exactly once and is always the last event in the stream.
Rolling out selective 4-bit weight-only quantization for supported models, targeting non-sensitive layers...
Rolling out selective 4-bit weight-only quantization for supported models, targeting non-sensitive layers while preserving full precision for sensitive layers, in line with industry standards. Learn more in Supported Models.
Temperature Parameter: Increased Maximum The maximum allowed value for the temperature parameter for Chat...
Temperature Parameter: Increased Maximum The maximum allowed value for the temperature parameter for Chat Completions has been increased from 1.5 to 2.0. This aligns with the range supported by OpenAI, Fireworks, and other major LLM providers. Higher temperature values produce more random and creative outputs.