From 19b7dac7190ef83af516269375b7ce8961022896 Mon Sep 17 00:00:00 2001 From: Seb Duerr Date: Mon, 29 Dec 2025 19:56:59 -0800 Subject: [PATCH] fix: update Cerebras maxTokens to 16384 (#10387) --- packages/types/src/providers/cerebras.ts | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) diff --git a/packages/types/src/providers/cerebras.ts b/packages/types/src/providers/cerebras.ts index 54b314b6db..8e0c2f9413 100644 --- a/packages/types/src/providers/cerebras.ts +++ b/packages/types/src/providers/cerebras.ts @@ -7,7 +7,7 @@ export const cerebrasDefaultModelId: CerebrasModelId = "gpt-oss-120b" export const cerebrasModels = { "zai-glm-4.6": { - maxTokens: 8192, // Conservative default to avoid premature rate limiting (Cerebras reserves quota upfront) + maxTokens: 16384, // Conservative default to avoid premature rate limiting (Cerebras reserves quota upfront) contextWindow: 131072, supportsImages: false, supportsPromptCache: false, @@ -18,7 +18,7 @@ export const cerebrasModels = { description: "Highly intelligent general purpose model with up to 1,000 tokens/s", }, "qwen-3-235b-a22b-instruct-2507": { - maxTokens: 8192, // Conservative default to avoid premature rate limiting + maxTokens: 16384, // Conservative default to avoid premature rate limiting contextWindow: 64000, supportsImages: false, supportsPromptCache: false, @@ -29,7 +29,7 @@ export const cerebrasModels = { description: "Intelligent model with ~1400 tokens/s", }, "llama-3.3-70b": { - maxTokens: 8192, // Conservative default to avoid premature rate limiting + maxTokens: 16384, // Conservative default to avoid premature rate limiting contextWindow: 64000, supportsImages: false, supportsPromptCache: false, @@ -40,7 +40,7 @@ export const cerebrasModels = { description: "Powerful model with ~2600 tokens/s", }, "qwen-3-32b": { - maxTokens: 8192, // Conservative default to avoid premature rate limiting + maxTokens: 16384, // Conservative default to avoid premature rate limiting contextWindow: 64000, supportsImages: false, supportsPromptCache: false, @@ -51,7 +51,7 @@ export const cerebrasModels = { description: "SOTA coding performance with ~2500 tokens/s", }, "gpt-oss-120b": { - maxTokens: 8192, // Conservative default to avoid premature rate limiting + maxTokens: 16384, // Conservative default to avoid premature rate limiting contextWindow: 64000, supportsImages: false, supportsPromptCache: false,