From 72db98a0f6ac6f1586eeaa20e51985ce2c65eba1 Mon Sep 17 00:00:00 2001
From: erin2722 <16248113+erin2722@users.noreply.github.com>
Date: Sun, 5 Jul 2026 10:48:27 +0000
Subject: [PATCH 1/2] chore: update model catalog from bot issues

---
 packages/proxy/schema/index.ts        |  3 +++
 packages/proxy/schema/model_list.json | 19 +++++++++++++++----
 2 files changed, 18 insertions(+), 4 deletions(-)

diff --git a/packages/proxy/schema/index.ts b/packages/proxy/schema/index.ts
index 906f5265..d83bb292 100644
--- a/packages/proxy/schema/index.ts
+++ b/packages/proxy/schema/index.ts
@@ -1062,6 +1062,9 @@ export const AvailableEndpointTypes: { [name: string]: ModelEndpointType[] } = {
   "grok-latest": ["xAI"],
   "gemma-4-31b": ["cerebras"],
   "gpt-5.4-mini-2026-03-17": ["openai", "azure"],
+  "pearl-ai/gemma-4-31b-it": ["together"],
+  "gpt-5-mini": ["openai", "azure"],
+  "gpt-5-nano": ["openai", "azure"],
 };
 
 const modelEndpointAvailableModels = getAvailableModels();
diff --git a/packages/proxy/schema/model_list.json b/packages/proxy/schema/model_list.json
index 8b07db8f..d7b893ec 100644
--- a/packages/proxy/schema/model_list.json
+++ b/packages/proxy/schema/model_list.json
@@ -534,7 +534,7 @@
     "input_cache_read_cost_per_mil_tokens": 0.125,
     "displayName": "GPT-5",
     "reasoning": true,
-    "max_input_tokens": 272000,
+    "max_input_tokens": 400000,
     "max_output_tokens": 128000,
     "available_providers": [
       "openai",
@@ -584,7 +584,7 @@
     "input_cache_read_cost_per_mil_tokens": 0.025,
     "displayName": "GPT-5 mini",
     "reasoning": true,
-    "max_input_tokens": 272000,
+    "max_input_tokens": 400000,
     "max_output_tokens": 128000,
     "available_providers": [
       "openai",
@@ -600,7 +600,7 @@
     "input_cache_read_cost_per_mil_tokens": 0.005,
     "displayName": "GPT-5 nano",
     "reasoning": true,
-    "max_input_tokens": 272000,
+    "max_input_tokens": 400000,
     "max_output_tokens": 128000,
     "available_providers": [
       "openai",
@@ -615,7 +615,7 @@
     "output_cost_per_mil_tokens": 120,
     "displayName": "GPT-5 Pro",
     "reasoning": true,
-    "max_input_tokens": 128000,
+    "max_input_tokens": 400000,
     "max_output_tokens": 272000,
     "available_providers": [
       "openai",
@@ -12323,5 +12323,16 @@
     "available_providers": [
       "cerebras"
     ]
+  },
+  "pearl-ai/gemma-4-31b-it": {
+    "format": "openai",
+    "flavor": "chat",
+    "input_cost_per_mil_tokens": 0.28,
+    "output_cost_per_mil_tokens": 0.86,
+    "displayName": "Gemma 4 31B IT (Pearl)",
+    "max_input_tokens": 32000,
+    "available_providers": [
+      "together"
+    ]
   }
 }

From 71544ff72292e8bd7296206fde4c181fc28370df Mon Sep 17 00:00:00 2001
From: Erin McNulty <erin.mcnulty@braintrustdata.com>
Date: Mon, 6 Jul 2026 14:08:01 +0000
Subject: [PATCH 2/2] fix: keep gpt-5 max_input_tokens at the input cap, not
 the context window

#915/#916 changed gpt-5, gpt-5-mini, gpt-5-nano max_input_tokens 272000->400000
and gpt-5-pro 128000->400000, but 400000 is the total CONTEXT window, not the
accepted input size. The catalog's max_input_tokens is the input cap: for the
gpt-5 family max_input + max_output = 400000 (gpt-5/mini/nano: 272000 + 128000;
gpt-5-pro: 128000 + 272000). OpenAI's model page confirms gpt-5 = 400k context /
128k output. Revert to the correct per-model input caps (Codex P1).

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
---
 packages/proxy/schema/model_list.json | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/packages/proxy/schema/model_list.json b/packages/proxy/schema/model_list.json
index d7b893ec..a11dff89 100644
--- a/packages/proxy/schema/model_list.json
+++ b/packages/proxy/schema/model_list.json
@@ -534,7 +534,7 @@
     "input_cache_read_cost_per_mil_tokens": 0.125,
     "displayName": "GPT-5",
     "reasoning": true,
-    "max_input_tokens": 400000,
+    "max_input_tokens": 272000,
     "max_output_tokens": 128000,
     "available_providers": [
       "openai",
@@ -584,7 +584,7 @@
     "input_cache_read_cost_per_mil_tokens": 0.025,
     "displayName": "GPT-5 mini",
     "reasoning": true,
-    "max_input_tokens": 400000,
+    "max_input_tokens": 272000,
     "max_output_tokens": 128000,
     "available_providers": [
       "openai",
@@ -600,7 +600,7 @@
     "input_cache_read_cost_per_mil_tokens": 0.005,
     "displayName": "GPT-5 nano",
     "reasoning": true,
-    "max_input_tokens": 400000,
+    "max_input_tokens": 272000,
     "max_output_tokens": 128000,
     "available_providers": [
       "openai",
@@ -615,7 +615,7 @@
     "output_cost_per_mil_tokens": 120,
     "displayName": "GPT-5 Pro",
     "reasoning": true,
-    "max_input_tokens": 400000,
+    "max_input_tokens": 128000,
     "max_output_tokens": 272000,
     "available_providers": [
       "openai",