From 84ec4464933380238a22348349405d5cda21df0b Mon Sep 17 00:00:00 2001
From: AdithyaJob <adithya.job@runpod.io>
Date: Wed, 17 Jun 2026 22:19:58 -0700
Subject: [PATCH] added configs for  Gemma 4 31B and GPT-OSS 120B

---
 configs/gemma/gemma4_31b_it.yaml  | 11 +++++++++++
 configs/gpt-oss/gpt_oss_120b.yaml |  9 +++++++++
 2 files changed, 20 insertions(+)
 create mode 100644 configs/gemma/gemma4_31b_it.yaml
 create mode 100644 configs/gpt-oss/gpt_oss_120b.yaml

diff --git a/configs/gemma/gemma4_31b_it.yaml b/configs/gemma/gemma4_31b_it.yaml
new file mode 100644
index 0000000..eaab1c3
--- /dev/null
+++ b/configs/gemma/gemma4_31b_it.yaml
@@ -0,0 +1,11 @@
+model: google/gemma-4-31b-it
+gpu-memory-utilization: 0.95
+max-model-len: 8192
+dtype: auto
+trust-remote-code: true
+quantization: fp8
+kv-cache-dtype: fp8
+enforce-eager: false
+enable-prefix-caching: true
+enable-chunked-prefill: true
+speculative-config: '{"model":"RedHatAI/gemma-4-31B-it-speculator.eagle3","method":"eagle3","num_speculative_tokens":3}'
diff --git a/configs/gpt-oss/gpt_oss_120b.yaml b/configs/gpt-oss/gpt_oss_120b.yaml
new file mode 100644
index 0000000..8ef2fbd
--- /dev/null
+++ b/configs/gpt-oss/gpt_oss_120b.yaml
@@ -0,0 +1,9 @@
+model: openai/gpt-oss-120b
+gpu-memory-utilization: 0.95
+max-model-len: 8192
+dtype: auto
+trust-remote-code: true
+enforce-eager: false
+enable-prefix-caching: true
+enable-chunked-prefill: true
+speculative-config: '{"model":"RedHatAI/gpt-oss-120b-speculator.eagle3","method":"eagle3","num_speculative_tokens":3}'