Releasesby GlyphAdded support for retrieving log probabilities via the logprobs parameter for the Completions endpoint
Added support for retrieving log probabilities via the logprobs parameter for the Completions endpoint. Set to an integer (0-20) to return the most likely alternative tokens at each position, or null to disable.
Also shipped Oct 22, 2025Cerebras in October 2025 →
OTLP Ingestion Endpoints for Traces and Logs are in Open Beta
You can send your OTLP traces and logs directly to Sentry using our new ingestion endpoints
SolarWinds Enterprise data source for Grafana
Introducing SolarWinds data source plugin. SolarWinds is a network and infrastructure monitoring platform that provides tools to monitor, manage network performance, server health, and application availability.
Modal 1.2.1
- It's now possible to override the default -dev suffix applied to the autogenerated URLs for ephemeral Apps (i.e., when using modal serve) via a new dev_suffix field in the .modal.toml config file, or equivalently with the MODAL_DEV_SUFFIX environment variable.
Prisma ORM 6.18.0, Postgres Database Metrics and Config Modernization
Prisma ORM 6.18.0 includes updates for database metrics and configuration modernization for Prisma Postgres.
Sentry Ruby SDK 6.0.0
The new SDK allows ignoring status codes for Performance, setting a custom sampling interval for profiling and drops the async option.
Switch accounts easily on the mobile app
With new multi-account support on mobile, quickly move between accounts without logging out or losing progress.
More from Cerebras InferenceEvery Cerebras Inference release →
Deprecated qwen-3-coder-480b
Deprecated qwen-3-coder-480b. We recommend migrating to Z.ai GLM 4.7.
We're gradually introducing multi-token streaming across our models
We're gradually introducing multi-token streaming across our models. Instead of sending tokens individually, we'll now deliver them in batches through 200 evenly-spaced events per second. This change eliminates the artificial delays that occurred with single-token streaming.
Updated support for OpenAI GPT-OSS (gpt-oss-120b) - Tool calling with strict: true (constrained decoding) -...
Updated support for OpenAI GPT-OSS (gpt-oss-120b) Tool calling with strict: true (constrained decoding) Response format with json_schema when using strict=true Response format with json_object when using strict=true tool_choice: none Updated JSON Schema limitations: Nested levels increases from 5 to 10 Max properties increases from 100 to 500 Enabled...
We now support Predicted Outputs for faster generation when large portions of the output are known in advance
We now support Predicted Outputs for faster generation when large portions of the output are known in advance. Provide a draft of the expected response via the prediction field, and the model will efficiently reuse unchanged tokens while regenerating only those that differ. Ideal for code editing, document revisions, and template-based responses.