From c60f14e00233ddf6d73dbf5840eb1855df94e32a Mon Sep 17 00:00:00 2001 From: Ray Andrew Date: Mon, 16 Feb 2026 13:09:51 -0600 Subject: [PATCH] refactor(gateway): reorder and restructure llama-3.3-70b model routes in configuration --- llm-gateway.yaml | 24 ++++++++++++------------ 1 file changed, 12 insertions(+), 12 deletions(-) diff --git a/llm-gateway.yaml b/llm-gateway.yaml index b6a897b..4025d81 100644 --- a/llm-gateway.yaml +++ b/llm-gateway.yaml @@ -117,18 +117,6 @@ models: # model: "stepfun/step-3.5-flash:free" # ═══ TIER 2: Low cost (Groq, Cerebras — free tier with rate limits) ═══ - - name: "llama-3.3-70b" - routes: - - provider: groq - model: "llama-3.3-70b-versatile" - pricing: { input: 0.59, output: 0.79 } - - provider: deepinfra - model: "meta-llama/Llama-3.3-70B-Instruct-Turbo" - pricing: { input: 0.23, output: 0.40 } - - provider: cerebras - model: "llama-3.3-70b" - pricing: { input: 0.85, output: 1.20 } - - name: "llama-3.1-8b" routes: - provider: groq @@ -141,6 +129,18 @@ models: model: "meta-llama/Meta-Llama-3.1-8B-Instruct" pricing: { input: 0.03, output: 0.05 } + - name: "llama-3.3-70b" + routes: + - provider: deepinfra + model: "meta-llama/Llama-3.3-70B-Instruct-Turbo" + pricing: { input: 0.23, output: 0.40 } + - provider: groq + model: "llama-3.3-70b-versatile" + pricing: { input: 0.59, output: 0.79 } + - provider: cerebras + model: "llama-3.3-70b" + pricing: { input: 0.85, output: 1.20 } + - name: "gpt-oss" routes: - provider: groq