'%3e%3crect%20x='16'%20y='51'%20width='68'%20height='27'%20rx='13.5'/%3e%3ccircle%20cx='32'%20cy='53'%20r='15'/%3e%3ccircle%20cx='53'%20cy='43'%20r='20'/%3e%3ccircle%20cx='71'%20cy='57'%20r='12'/%3e%3c/g%3e%3c/svg%3e){kind=small}
Cloud Changelog
Cerebras Inference changelog 2025
26 announcements, 2025
All Cerebras Inference changes · 62 announcements since February 2025
Free. Unsubscribe any time.
RSS feed for Cerebras Inference
December 20254
Batch API and Files API: up to 50,000 requests per batch at 50% lower cost
Cerebras Inference adds parallel tool calling for multiple tool invocations per response
Streaming supported for all models, plus reasoning_format parameter for reasoning models
Cerebras Inference adds prompt caching to reduce latency and time to first token
November 20255
Cerebras Inference supports Predicted Outputs via the prediction field
Cerebras Inference deprecation notice recommends migrating to GPT OSS 120B
Z.ai GLM 4.6 (zai-glm-4.6) available in preview on Cerebras Inference
qwen-3-coder-480b deprecated on Cerebras Inference, migrate to Z.ai GLM 4.7
Cerebras Inference deprecations announced, see Deprecations page for details
October 20254
August 20253
July 20256
Qwen 3 Coder 480B (qwen-3-coder-480b) in preview on Cerebras Inference
Qwen 3 235B Thinking (qwen-3-235b-a22b-thinking-2507) in preview on Cerebras Inference
Migrate to Qwen 3 235B Instruct or Thinking (coming soon) as deprecation notice
Qwen 3 235B Instruct in preview as qwen-3-235b-a22b-instruct-2507
Llama 4 Maverick in preview as llama-4-maverick-17b-128e-instruct